anaesthesia for the high risk patient - i. mcconachie (2002) ww

ANAESTHESIA FOR THE HIGH RISK PATIENT

Prelims.qxd 2/7/02 10:10 AM Page i

Prelims.qxd 2/7/02 10:10 AM Page ii

This Page Intentionally Left Blank


Edited byDr Ian McConachie, Consultant in Anaesthesia & Intensive Care

Blackpool Victoria Hospital

Prelims.qxd 2/7/02 10:10 AM Page iii

© 2002

Greenwich Medical Media Limited137 Euston Road

London NW1 2AA

870 Market Street, Ste 720,San Francisco, CA 94102

ISBN 1 84110 072 2

First Published 2002

Apart from any fair dealing for the purposes of research or private study, or criticismor review, as permitted under the UK Copyright Designs and Patents Act 1988, thispublication may not be reproduced, stored, or transmitted, in any form or by anymeans, without the prior permission in writing of the publishers, or in the case ofreprographic reproduction only in accordance with the terms of the licences issuedby the appropriate Reproduction Rights Organisations outside the UK. Enquiriesconcerning reproduction outside the terms stated here should be sent to the publish-ers at the London address printed above.

The right of Ian McConachie to be identified as editor of this work has been assertedby him in accordance with the Copyright Designs and Patents Act 1988.

While the advice and information in this book is believed to be true and accurate,neither the authors nor the publisher can accept responsibility or liability for anyloss or damage arising from actions or decisions based in this book. The ultimateresponsibility for treatment of patients and the interpretation lies with the medicalpractitioner. The opinions expressed are those of the author and the inclusion inthe book of information relating to a particular product, method or technique doesnot amount to an endorsement of its value or quality, or of the claims made of it byits manufacturers. Every effort has been made to check drug dosages; however, it isstill possible that errors have occured. Furthermore, dosage schedules are constantlybeing revised and new side effects recognised. For these reasons, the medical prac-titioners are strongly urged to consult the drug companies’ printed instructionsbefore administering any of the drugs mentioned in this book.

The publisher makes no representation,express or implied,with regard to the accuracyof the information contained in this book and cannot accept any legal responsibilityor liability for any errors or omissions that may be made.

A catalogue record for this book is available from the British Library.

Project ManagerGavin Smith

Typeset by Charon Tec Pvt. Ltd, Chennai, India

Printed by Ashford Colour Press Ltd, Hants

Distributed by Plymbridge Distributors Ltd and in the USA by Jamco Distribution

Visit our website at www.greenwich-medical.co.uk

Prelims.qxd 2/7/02 10:10 AM Page iv

CONTENTS

Preface ........................................................................................................ vii

Contributors .............................................................................................. ix

1. Epidemiology and identification of the high-risk surgical patient .......... 1A. Adams

2. Respiratory risk and complications ...................................................... 29A. Adams

3. Lessons from the National Confidential Enquiry into Perioperative Deaths ............................................................................ 41K. Paramesh and C. Dunkley

4. Analgesia for the high risk patient ........................................................ 51F. Duncan and D.J. Counsell

5. Local anaesthetic techniques ................................................................ 65B. Lord

6. The critically ill patient in the operating theatre .................................. 77D. Hume and I. McConachie

7. The elderly patient .............................................................................. 101S. Vaughan

8. Perioperative optimisation .................................................................. 117M. Cutts

9. The patient with coronary heart disease .............................................. 127S. Lakshmanan and M. Hartley

v

Prelims.qxd 2/7/02 10:10 AM Page v

10. Valvular heart disease and pulmonary hypertension .............................. 141C. Harle

11. Emergency abdominal aortic surgery .................................................... 153G. Johnson and M. Chamberlain

12. Gastrointestinal surgery ........................................................................ 165A. Heard and N. Harper

13. Perioperative renal insufficiency and failure .......................................... 179I. McConachie

14. The role of the cardiology consult ........................................................ 199S. Bulugahapitiya and D. Hesketh Roberts

15. The risk of anaemia and blood transfusion ............................................ 215M. Bewsher

16. Admission criteria for HDU and ICU .................................................. 227V. Prasad and J. Cupitt

17. The meaning of risk ............................................................................ 239A. Adams

Index .......................................................................................................... 249

CONTENTS

vi

Prelims.qxd 2/7/02 10:10 AM Page vi

PREFACE

This text:

• Is aimed primarily at trainees in Anaesthesia though more experiencedpractitioners may find it useful as a refresher in recent concepts andadvances. A basic knowledge of physiology, pharmacology and anaes-thesia is assumed.

• May be a useful ‘aide memoire’ for the FRCA and other examinations inanaesthesia.

• Aims to provide practical information on the management of high-riskpatients presenting for surgery as well as sufficient background informa-tion to enable understanding of the principles and rationale behindtheir anaesthetic and perioperative management. We hope it will proveuseful but we would emphasise that this, or any other book, is no substitute for experienced supervision, support and training.

• Is not a substitute for the major anaesthetic texts but concentrates onprinciples of management of the most challenging anaesthetic cases.

• Aims to provide guidance to help manage these patients in the peri-operative period in line with modern concepts of critical care ‘outreach’and the potential role of the anaesthetist as perioperative physician.

• Emphasises cardiovascular risk, cardiac disease and cardiac managementas these are undoubtedly the most important aspects of perioperativeanaesthetic risk.

• The choice of topics is, by nature of the size of text intended, selectivebut should appeal and be useful to the majority of practitioners. Importantinformation not readily available in similar texts, e.g. a summary of allconfidential enquiry into perioperative deaths (CEPOD) reports, peri-operative renal failure, the role of the cardiology consult and indications

vii

Prelims.qxd 2/7/02 10:10 AM Page vii

for admission to high dependency unit (HDU) and intensive care unit(ICU) are included.

• The format is designed to provide easy access to information presentedin a concise manner. We have tried to eliminate all superfluous mate-rial. Selected important or controversial references are presented as wellas suggestions for Further reading. The style of the chapters vary. Thisis deliberate. Some relate more to basic principles, physiology, pharma-cology, etc. – bookwork. Others are more practical in nature, discussingthe principles of anaesthetic techniques for certain high-risk situations.

• The authors are all experienced practitioners working in a large, busyDGH with a high proportion of sick, elderly patients presenting forboth elective and emergency surgery. The authors are committed toproviding a high level of perioperative care of patients undergoinganaesthesia. We make no apologies for repetition of important principlesand facts.

I McConachieBlackpool

2001

PREFACE

viii

Prelims.qxd 2/7/02 10:10 AM Page viii

CONTRIBUTORS

Blackpool Victoria Hospital, Whinney Heys Road,Blackpool FY3 8NR, England

Department of AnaesthesiaDr A. Adams FRCS FRCADr K. Paramesh FRCADr C. Dunkley FRCADr D.J. Counsell FRCADr B. Lord FRCADr D. Hume FRCADr S. Vaughan MRCP FRCADr M. Cutts MRCP FRCADr S. Lakshmanan FRCADr M. Hartley FRCADr C. Harle FRCADr G. Johnson FRCADr M. Chamberlain FRCADr A. Heard FRCADr N. Harper FRCADr V. Prasad FRCADr J. Cupitt FRCADr M. Bewsher FRCADr I. McConachie FRCA

Department of CardiologyDr S. Bulugahapitiya MRCPDr D. Hesketh Roberts FRCP

Acute Pain ServiceMs F. Duncan SRN

ix

Prelims.qxd 2/7/02 10:10 AM Page ix

Prelims.qxd 2/7/02 10:10 AM Page x


1

1EPIDEMIOLOGY AND IDENTIFICATION

OF THE HIGH-RISK SURGICALPATIENT

Patients that present for surgery may be at increased clinical risk for a variety ofreasons. These reasons can be broadly divided into the following categories.

Availability of appropriately experienced staff

The confidential enquiry into peri-operative deaths (CEPOD) has identified theimportance of training and adequate experience for medical staff. Training and,therefore, competence is an issue:

• ‘Board-certified’ trauma surgeons improve outcome following majortrauma.1

• Trained specialists improve the outcome of septic shock in intensivecare unit (ICU).2

Clinical volume for surgeons may be important:

• Studies have looked at both hospital volume and individual surgeonsvolume. Both may be important (with some evidence that ‘high volume’ hospital may compensate up to a point for ‘low volume’ surgeons).3

• Several studies have looked at bowel cancer surgery. There are signifi-cant differences in outcome between surgeons for colon cancer surgery.It has been suggested that such surgery should only be undertaken byspecialist surgeons. One may assume that volume would be a factor inthis and studies support this.

• However, one must be wary of uncritically accepting this – there aredifferences in surgical practice between the US and UK surgeons.

Chap-01.qxd 2/1/02 12:03 PM Page 1

For example, in a paper showing importance of volume in colorectal surgery,3 low volume was five or less cases in a year. High volume was� 10 cases a year (and these surgeons were in a minority). Few UKgeneral surgeons would, therefore, not fall in the high volume group –with some performing that many in a month.

As regards, the anaesthetist, there have been few studies which have effectivelycome down to assessing the role of the competence of the anaesthetist on risk andoutcome.

• One study of patients undergoing coronary artery surgery found thatthe only non-patient related factors influencing outcome were cardiacbypass time and the anaesthetist.4

One can expect more such studies in the future.

Timing of surgery

CEPOD has confirmed that surgery performed at night, when staff are morelikely to be fatigued, is more hazardous and contributes to increased mortality.5

Availability of equipment

It is clear how the absence of basic equipment (e.g. capnography or pulse oxi-metry) might contribute to increased risk.

Patient factors

Many of these (see below) may be beyond the control or influence of the clin-icians but may still be associated with increased risk or worse outcome.

Gender

The influence of gender on cardiovascular risk is discussed below. Some studieshave investigated the role of gender in peri-operative risk and surgical risk andoutcome:

• Females have significantly better outcomes including mortality andrecurrence rates from melanomas.6

• The incidence of septic shock requiring intensive care is significantlyless in females.7 No differences in outcome, however, were demon-strated.

• Aligned with this is the observation that males have a higher incidenceof infection following trauma.8


2

Chap-01.qxd 2/1/02 12:03 PM Page 2

• Females have a worse outcome from IPPV but this was less important inpredicting outcome than age,APACHE scores or presence of ARDS.9

• Females have a worse outcome following vascular surgery.10

Although gender may influence risks and outcome, this must be put into perspec-tive and is only believed to be a minor risk factor overall. Vascular surgery may bean exception in that several studies suggest gender to be an important risk factor.

Age

Discussed as a cardiovascular risk factor below and also in the chapter on the elderly patient.

Race

The influence of a patient’s race on risk and outcome is poorly understood and isa very sensitive issue – not least because of concerns that any such differences mayreflect prejudice or access to health care. Differences in ethnic incidence and drugresponses in hypertension have long been recognised. A few studies have examinedrace as a factor in surgical and peri-operative risk and outcome:

• Prostate cancer may be intrinsically more aggressive with a worse out-come in North American negroes.11

• There are similar results for endometrial cancer.12

Race has not been identified as an anaesthetic risk factor.

Genetic predisposition

The understanding of genetic predispositions to risk of sepsis and cardiac progno-sis is still in its infancy. No work has been done on surgical outcomes but a geneticpredisposition to high levels of angiotensin converting enzyme is associated withreduced survival following diagnosis of cardiac failure.13 This may have implica-tions for cardiac reserve and response to physiological stress peri-operatively. It isalso very likely that the inflammatory response and response to infection is, in part,genetically predetermined.

Clinical conditions

There are numerous examples of high profile clinical conditions that readily pre-dict high peri-operative risk:

• leaking abdominal aortic aneurysm,

• an unstarved patient with difficult intubation for emergency surgery,

• the emergency obstetric patient for caesarean section,

EPIDEMIOLOGY AND IDENTIFICATION OF THE HIGH-RISK SURGICAL PATIENT

3

Chap-01.qxd 2/1/02 12:03 PM Page 3

• fractured neck of femur,

• myopathic conditions,

• malignant hyperthermia,

• hereditary mastocystosis,

• latex allergy.

Many of these conditions are rare and would account for a small fraction of peri-operative deaths whilst others represent conditions that predispose to increasedmortality for multifactorial reasons.

The majority of this chapter discusses cardiovascular disease as the most importantfactor in risk.

THE SIGNIFICANCE OF CARDIOVASCULAR DISEASE

There are currently two main theories as to how cardiac disease might contributeto peri-operative mortality in the surgical patient:

• Myocardial ischaemia: Tachycardia and increased myocardial oxygendemand increases shear stress on atherosclerotic plaques, which leads toplaque rupture, coronary thrombosis and myocardial infarction (MI).

• Poor cardiopulmonary physiological reserve: The physiologicalreserve of the heart and lungs is insufficient to meet the increaseddemands of surgery. In physiological terms, oxygen delivery does notfulfil oxygen consumption requirements. End-organ ischaemia resultsin multi-organ dysfunction syndrome (MODS) and death.

Interestingly, with respect to this second hypothesis, it has recently been shownthat a pre-operative intramucosal gastric pH of � 7.35 predicted increased mor-tality.14 pHi is a marker of blood supply to the stomach, and low values are thoughtto reflect inadequate oxygen delivery to the gut.

The second hypothesis also explains the importance of adequate pulmonary reserveand the contribution of pulmonary disease to surgical mortality in the peri-operative period. Pulmonary risk stratification is discussed in the next chapter.

RISK STRATIFICATION

The grading of patients into incremental levels of risk is known as risk stratifica-tion. There are a number of reasons why it is useful to identify who is at high risk:

• to identify those suitable for coronary revascularisation (either bypassgrafting or angioplasty),


4

Chap-01.qxd 2/1/02 12:03 PM Page 4

• to identify who would benefit from other peri-operative risk-reductionstrategies.

Both strategies are likely to have major resource implications, however, the latteris increasingly being recognised as that most likely to improve outcome.15

CLINICAL FACTORS ASSOCIATED WITH INCREASED CARDIAC RISK

Advanced age (see also Chapter 7)

• Elderly patients have shorter life expectancy.

• Elderly patients have higher rates of treatment-related risks.

• Age increases the likelihood of coronary artery disease (CAD).

• The mortality of acute MI increases dramatically in the aged.

• Intra-operative or peri-operative MI has a higher mortality in the aged.

• CEPOD data shows for deaths within 30 days of surgery the peak ageis 70–74 for males and 80–84 for females.

• In some elderly patients the risks of surgery may come close to risks ofdoing nothing.

Gender

• Premenopausal women have a lower incidence of CAD.

• CAD occurs 10 or more years later in women than in men.16

• Diabetic women have an increased risk, which is equivalent to men ofthe same age.

• The mortality rate following acute MI is greater for women than formen, but older age and diabetes mellitus account for much of this difference.17

Coronary artery disease

• A previous history of acute MI, bypass grafting, coronary angioplasty, orcoronary angiography demonstrating coronary stenosis are all obviousindicators of ongoing CAD.

• Patients with a prior history of MI have an increased risk of peri-operative MI that is graded according to the time interval since theirinfarction (table 1.1).18

• The difficulty arises in identifying those patients with occult CAD.


5

Chap-01.qxd 2/1/02 12:03 PM Page 5

• In some patients symptoms may not occur due to functional limitationby arthritis or peripheral vascular disease.

It is these patients that may benefit from non-invasive testing to determine:

• the amount of myocardium critically perfused,

• the amount of stress required to produce ischaemia (ischaemic threshold),

• the ventricular function.

Non-invasive testing should not be performed in patients unsuitable for myocar-dial revascularisation.

• In those patients where the risk of coronary revascularisation is greaterthan the risk of non-cardiac surgery, the issue is whether to proceedwith non-cardiac surgery anyway.

Hypertension

• Moderate hypertension is not an independent risk factor for peri-operative cardiovascular complications. There is thus, no need to delaysurgery for mild or moderate hypertension with no associated meta-bolic or cardiovascular abnormalities.19

• However, as hypertension is associated with ischaemic heart disease, itshould raise the index of suspicion.

• Poorly controlled hypertension can cause exaggerated intra-operativeblood pressure variation and electrocardiograph (ECG) evidence ofmyocardial ischaemia, which has been shown to be a factor in post-operative cardiac morbidity.

• Effective pre-operative blood pressure control has been shown toreduce the incidence of peri-operative ischaemia so all antihyper-tensive medications should be continued during the peri-operativeperiod.17,20

• Severe hypertension (diastolic � 110 mmHg) should be controlledbefore surgery when possible.


6

Table 1.1 – Peri-operative infarction rates following a recent MI.18

Time since MI Rate of new infarct (%)

� 6 months 5Between 3 and 6 months 15� 6 months 37

Chap-01.qxd 2/1/02 12:03 PM Page 6

• Any decision to delay surgery for hypertension should balance theurgency of surgery against the potential benefit of aggressive medicaloptimisation.

• If urgent surgery is essential, intravenous � blockade can rapidly achieveeffective control and reduce the number and duration of peri-operativecoronary ischaemic episodes.

Congestive heart failure

• Congestive heart failure (CHF) has been identified in numerous stud-ies as a predictor of a poor outcome in non-cardiac surgery.

• Validated predictive clinical signs include the presence of a third heartsound, and bibasal crackles.

• In Goldman’s study,21 the presence of a third heart sound or signs ofCHF were associated with a substantially increased risk during non-cardiac surgery.

• Consideration of the aetiology of heart failure is particularly important;heart failure of ischaemic origin carries a greater significance and riskthan that caused by hypertension.19

Valvular heart disease

With all murmurs one needs to determine:

• if the murmur is organic or simply a flow murmur,

• if the murmur is significant,

• whether endocarditis prophylaxis is required,

• the severity of the valvular lesion,

• if an echo is indicated.

Symptomatic stenotic lesions

• Symptomatic stenotic lesions are associated with severe peri-operativeCHF or shock and normally require percutaneous valvotomy or valvereplacement prior to surgery to reduce cardiac risk.

• Severe aortic stenosis poses the greatest risk and elective non-cardiacsurgery should generally be postponed until fully assessed.21

• Mitral stenosis, although rare, increases the risk of CHF and balloonvalvuloplasty or open repair may reduce peri-operative risk.22


7

Chap-01.qxd 2/1/02 12:03 PM Page 7

• Mild or moderate stenosis requires careful avoidance of tachycardia tominimise the reduction in diastolic filling time that can precipitatesevere pulmonary congestion.

Symptomatic regurgitant valve disease

• Symptomatic regurgitant valve disease is usually better tolerated peri-operatively and may be stabilised pre-operatively with intensive medicaltherapy and monitoring.

• Usually treated definitively with valve repair or replacement either afteror at the same time as non-cardiac surgery if left ventricular (LV) func-tion is adequate.

• Aortic regurgitation requires attention to volume control and afterloadreduction.

• In severe aortic regurgitation slow heart rates increase the volume ofregurgitation by increasing the amount of time in diastole.

• Moderate tachycardia is preferred as faster heart rates reduce diastoleand, therefore, reduces the time for regurgitation.

Arrhythmias and conduction abnormalities

• Although both supraventricular and ventricular arrhythmias have beenidentified as independent risk factors for coronary events in the peri-operative period,21 they are probably significant because they reflect thepresence of underlying serious cardiopulmonary disease, drug toxicity,or metabolic abnormality.

Diabetes mellitus

• Diabetes mellitus increases both the likelihood and extent of CAD.

• Myocardial ischaemia and MI are more likely to be silent with diabetesmellitus.23

Peripheral vascular disease and cerebrovascular disease

• These conditions confer an increased risk for cardiac complications,since many of the risk factors contributing to peripheral vascular disease (diabetes mellitus, smoking, hyperlipidemia) are also risk factorsfor CAD.

• As mentioned above, symptoms suggestive of CAD may be masked byexercise limitations.


8

Chap-01.qxd 2/1/02 12:03 PM Page 8

• The presence of peripheral vascular disease is more important as a predictor of cardiac events than the actual vascular operation to be performed.19

SCORING SYSTEMS

Scoring systems are usually derived from the statistical analysis of large populationgroups where certain factors predictive of increased risk are identified andweighted according to their individual significance.

The ideal scoring system for predicting clinical risk would be:

• simple to use,

• highly sensitive,

• highly specific,

• high positive predictive value,

• cheap without requiring use of expensive resources or tests.

There are numerous scoring systems that have been published over the years, themost important of these include:

• American Society of Anesthesiologists ‘ASA’ Status,24

• Goldman’s Cardiac Risk Index,21

• Detsky’s Modified Cardiac Risk Index,25

• Lee’s Revised Cardiac Risk Index.26

American Society of Anesthesiologists ‘ASA’ Status

This classification of physical status was originally introduced in 1941 with sevenclasses, but was revised to its final form of just five classes in 196324 (table 1.2). Themain points to note regarding ASA status include:

• it stratifies patients by simple assessment of physical status,

• no expensive tests or clinical resources required,

• there can be considerable observer variability of patients’ physical status.

Important factors not taken into account include:

• age – some workers add an extra grade on for age � 75 years,

• complexity of operation,

• duration of operation,

• whether the disease process is incidental or factorial in current illness.


9

Chap-01.qxd 2/1/02 12:03 PM Page 9

Wolters et al.27 recently investigated whether the ASA classification and the pres-ence of peri-operative risk factors could be used as a predictive tool either for out-come or post-operative complications; factors shown to correlate with increasingclass of ASA included:

• intra-operative blood loss,

• duration of operation,

• duration of post-operative ventilation,

• post-operative wound and urinary tract infections,

• length of ICU stay and hospital stay,

• rates of pulmonary and cardiac complications,

• in-hospital mortality.

The variables found to be most important for predicting complications were ahigh ASA class, having a major operation, and having an emergency operation.

Many retrospective studies and a couple of prospective studies have demonstrateda correlation between ASA and peri-operative mortality and justifies the use ofASA classification as a crude predictor of patient outcome.

Goldman’s Cardiac Risk Index

This landmark paper was published in the NEJM 1977,21 and is a well-knownmethod for stratifying risk (table 1.3).

Its limitations are few but include:

• The index overestimated the incidence of cardiac morbidity in Class IVpatients undergoing non-cardiac surgery.

• The index underestimated risk in Class I and II patients undergoingaortic surgery.

• The study group included elective non-emergent cases only.


10

Table 1.2 – American Association of Anesthesiologists ‘ASA’ Status and mortality ranges foreach class.24

ASA class Definition Mortality (%)

I Healthy 0–0.3II Mild systemic disease with no functional limitation 0.3–1.4III Severe systemic disease with functional limitation 1.8–5.4IV Severe systemic disease – constant threat to life 7.8–25.9V Moribund patient unlikely to survive 24 h with or without operation 9.4–57.8E Suffix added to denote emergency operation

Chap-01.qxd 2/1/02 12:03 PM Page 10

Subsequently a number of authors attempted to improve upon the originalGoldman’s Cardiac Risk Index. In 1986 Detsky et al., published the ModifiedCardiac Risk Index25 in which a number of other clinical conditions were incorporated:

• Canadian cardiovascular society angina Classes III and IV,

• unstable angina,

• history of pulmonary oedema.

Since then other authors have suggested the addition of coronary perfusion scansor dobutamine stress echocardiography as a means of improving sensitivity andspecificity.

Revised Cardiac Risk Index

The Revised Cardiac Risk Index (table 1.4) is the most recent scoring systems andwas introduced by Lee et al. in 1999.26 The index was derived from a populationof 4000 patients, and identified the risk of major cardiac complications in a popu-lation undergoing major non-emergent non-cardiac surgery.


11

Table 1.3 – Goldman’s Cardiac Risk Index.21

Criteria Points

History Age � 70 5MI in previous 6 months 10

Examination S3 gallop or jugular venous distension 11Important valvular aortic stenosis 3

ECG Rhythm other than sinus or PACs 7� 5 PVCs/min at any time before operation 7

General status pO2 � 60 or pCO2 � 50 mmHg 3K � 3.0 or HCO3 � 20 mmol/lUrea � 18 mmol/l or Cr � 240 �molAbnormal AST (SGOT)Signs of chronic liver diseasePatient bedridden from non-cardiac causes

Operation Intraperitoneal, intrathoracic, or aortic 3emergency 4

Total possible points 53

Group Score Life-threatening Deaths (%)complications (%)

I 0–5 0.7 0.2II 6–12 5 1.5III 13–25 11 2.3IV 26–53 22 56

Chap-01.qxd 2/1/02 12:03 PM Page 11

Major cardiac complications included MI, pulmonary oedema, ventricular fibril-lation or primary cardiac arrest, or complete heart block.

Its advantages over the earlier scoring systems include:

• only six prognostic factors,

• simple variables,

• dependent on presence or absence of conditions rather than estimatingdisease severity,

• less reliance on clinical assessment and judgement,

• could easily be incorporated on pre-operative evaluation forms.

The shortcomings of the system are that:

• it is not applicable to emergency surgery,

• it is not applicable to lower-risk populations,


12

Table 1.4 – Revised Cardiac Risk Index.26

Risk factors Inclusion criteria

Ischaemic heart disease MIQ wavesAnginaNitratesPositive exercise stress test

CHF HistoryExaminationCXR

Cerebrovascular disease StrokeTIA

Insulin treated diabetes

Creatinine � 177 �mol

High-risk surgery AAA repairThoracicAbdominal

Revised Cardiac No. of factors Proportion of Major cardiac Risk Index population (%) complications (%)

Class I 0 36 0.4Class II 1 39 1.1Class III 2 18 4.6Class IV 3 or more 7 9.7

Patients with 0 or one risk factor accounted for 75% of population and had a risk of major cardiac event of 1.5%.Patients with two risk factors accounted for 18% of the population group with a risk of major cardiac event of 4.6%.Patients with three or more risk factors accounted for 7% of the population group with a risk of major cardiac event of 9.7%.

Chap-01.qxd 2/1/02 12:03 PM Page 12

• it may not be as reliable for pre-selected high-risk populations such aspatients undergoing major vascular surgery.

APACHE systems

APACHE is an acronym for Acute Physiology and Chronic Health Evaluation.APACHE II and III are scoring systems in widespread use in ICUs, but are unsuit-able as a pre-operative risk stratification tools because for score generation requires12 physiological parameters from the first 24 h of care as well as age and previoushealth status.

Possum

Possum is an acronym for the Physiological and Operative Severity Score for theEnumeration of Mortality and Morbidity. Copeland et al.28 developed this scor-ing system in 1991 for audit purposes. It requires 12 physiological variables, anumber of operative severity score factors and is reliant on outcome for final scoreand, therefore, is not suitable for pre-operative risk prediction:

• It has mainly been utilised in the UK to date.

• Its main use is to compare hospitals for audit purposes and identify dif-ferences between individual surgeons.

• The score may overpredict mortality in low-risk patients.

• It does not predict mortality accurately for ruptured aortic aneurysms.29

• Possum is better than APACHE II in predicting mortality in highdependency unit (HDU) patients.30

• For colorectal surgery, predicted mortality with Possum equals actualmortality.31

AMERICAN COLLEGE OF CARDIOLOGISTS/AMERICAN HEARTASSOCIATION GUIDELINES

In 1996 the American College of Cardiologists and the American HeartAssociation (ACC/AHA) introduced guidelines for the peri-operative cardiovas-cular evaluation of patients undergoing non-cardiac surgery:19

• The guidelines reflected the failings of the scoring systems and weredesigned to provide central,evidenced-based advice as a strategy to reducelitigation claims for suboptimal pre-operative management in the US.

• They were also an attempt to rationalise the increasing demand forexpensive risk stratification tests being requested as part of routinepre-operative assessments.


13

Chap-01.qxd 2/1/02 12:03 PM Page 13


14

Non-invasiverisk stratification

Invasive risk stratification(coronary angiography)

Angioplasty/stenting orbypass grafting

Other peri-operativerisk-reduction strategies

Clinicalpredictors

Minor Intermediate Major

Functionalcapacity Excellent Moderate Poor

Surgicalrisk Low Intermediate High

Definitive surgery

Figure 1.1 – Illustration of stepwise process to aid clinical risk assessment and stratification.19

The guidelines proposed a sequential stepwise strategy of risk assessment basedupon (figure 1.1):

• the identification of certain clinical predictors of risk,

• an assessment of the patients functional capacity,

• the type of surgery to be undertaken.

Chap-01.qxd 2/1/02 12:03 PM Page 14

Clinical predictors of coronary risk

Table 1.5 lists a number of clinical predictors of increasing risk for MI, CHF, anddeath established by several authors based on multivariate analysis:19

• Patients with minor predictors do not usually require any furthernon-invasive testing.

• Patients possessing intermediate predictors may be further stratifiedaccording to their functional capacity and surgery-specific risk.

• Major clinical predictors are so predictive of high risk that furtherstratification, using functional capacity and surgery-specific risk, is usu-ally unnecessary. These patients invariably require further non-invasiverisk stratification testing, and this often results in the delay or cancella-tion of all but emergency surgery.

Interestingly, the guidelines class a history of MI or pathological Q waves by ECGas an intermediate predictor, whereas a recent MI is a major predictor. They admitthat there are no adequate clinical trials on which to base firm recommendations,but suggest a delay of 4–6 weeks after MI before anaesthetising for elective surgery.

Functional capacity

• Assessment of an individual’s capacity to perform a number of differentphysical tasks has been shown to correlate with maximum oxygenuptake (vO2 max) by treadmill testing.

• Exercise tolerance or physical fitness can be assessed in metabolicequivalent levels, or ‘METs’, which is a validated method of determin-ing functional capacity from the patients history (table 1.6).


15

Table 1.5 – ACC AHA clinical predictors of coronary risk.19

Minor clinical predictors Intermediate predictors Major predictors

Advanced age Mild stable angina Class I/II Unstable coronary syndromesAbnormal ECG Previous MI by Hx or Q waves Recent MI � 30 days

LVH Compensated CCF Unstable or severe anginaLBBB IDDM Decompensated CCFST-T abnormalities Significant arrhythmias

Absence of sinus rhythm High grade AV blockLow functional capacity Symptomatic ventricularPrevious CVA arrhythmiasUncontrolled hypertension Supraventricular tachyarrhythmias

Severe valvular heart disease

Chap-01.qxd 2/1/02 12:03 PM Page 15

• Peri-operative and long-term cardiac risk is increased in patients unable toachieve four ‘MET’ levels of exercise during normal day-to-day living.32

• Poor functional capacity in patients with chronic CAD or in those convalescing after an acute cardiac event is associated with an increasedrisk of subsequent cardiac morbidity and mortality.

• Decreased functional capacity may be a result of several factors,including inadequate cardiac reserve, advanced age, transient myocardialdysfunction from myocardial ischaemia, deconditioning, and poor pulmonary reserve.

Surgical risk

In elective surgery the magnitude of cardiac risk has been stratified to the magni-tude of the surgical procedure and to major thoracic, abdominal,or vascular surgery,especially in patients of 70 years or older.19

In the ACC/AHA guidelines, surgical procedures are classified as low, intermedi-ate, or high risk (table 1.7) and are based upon the likelihood of non-fatal MI ordeath from cardiac causes:

• Low-risk procedures are usually short, with minimal fluid shifts, whilehigher-risk operations tend to be prolonged with large fluid shifts andgreater potential for post-operative myocardial ischaemia and respira-tory depression.

• Not surprisingly, major vascular procedures represent the highest-riskprocedures.


16

Table 1.6 – Metabolic equivalent levels with examples of common daily tasks.32

Metabolic levels Equivalent activity

1 MET EatDressUse toiletWalk indoors around houseLight house-workWalk on level ground at 2–3 mph

4 METs Climb flight of stairsWalk up hillRun a short distanceHeavy house-work, scrubbing floors, moving heavy furnitureWalk on level ground at 4 mphRecreational activity: golf, bowling, dancing, tennis

10 METs Strenuous sports: swimming, football, skiing, basketball

Chap-01.qxd 2/1/02 12:03 PM Page 16

• Superficial and ophthalmologic procedures represent the lowest riskand are rarely associated with excess morbidity and mortality.

• In the intermediate-risk category,morbidity and mortality vary,depend-ing on the surgical location and extent of the procedure.

• Emergency surgery is a special case as the necessity for immediate sur-gery precludes the time needed to fully evaluate and optimise thesepatients. The result is that cardiac complications are two to five times more likely to occur with emergency surgical procedures.33 Forexample, in asymptomatic elective abdominal aortic aneurysms repair,the mortality rate is 3.5% but this rises to 42% if ruptured.34

Individual clinician-based patient assessment

• This is a clinical philosophy that is gaining increasing acceptance as a valid alternative to the use of rigid risk scoring systems and guidelines.

• It relies on the history, physical examination and acquired clinical experi-ence to identify potential markers of increased risk.

• The difference between this approach and the use of scoring systems isthat clinicians have a unique ability to integrate the numerous otherfactors surrounding the patients’ presentation for surgery that con-tribute to peri-operative risk but would be impossible to quantify andvalidate in a scoring system.

• The result is a clinician-based assessment of risk that is specifically tailoredto each individual patient. Diagnostic testing and risk-reduction strate-gies are then selectively instigated, according to this experience, whereclinical and financial resources permit.

• Clinician-based judgement strategies are thus adaptable to individualclinical situations rather than rigidly applying an index score alonewhen estimating peri-operative risk.


17

Table 1.7 – Grade of risk (with reported cardiac risk) with type of surgical procedure. Risk is forcombined incidence of cardiac death and non-fatal MI.19

Low risk (��1%) Intermediate risk (��5%) High risk (��5%)

Endoscopic procedures Carotid endarterectomy Emergent major (particularly elderly)Superficial procedures Head and neck Aortic and other major vascular surgeryCataract surgery Intraperitoneal and intrathoracic Peripheral vascularBreast surgery Orthopaedic Anticipated prolonged surgical procedures

Prostate Associated with large fluid shifts/blood loss

Chap-01.qxd 2/1/02 12:03 PM Page 17

• This philosophy maintains that risk scoring systems and guidelines aremost useful, therefore, only as guides for inexperienced clinicians.

Non-invasive tests to stratify cardiovascular risk

The purpose of supplemental pre-operative testing is:

• to identify the presence of important pre-operative myocardial ischaemiaor cardiac arrhythmias,

• to estimate peri-operative cardiac risk and long-term prognosis,

• and to provide an objective measure of functional capacity.

They should not be performed if they are unlikely to influence management.

The ideal test for stratification should be:

• simple,

• cheap,

• non-invasive,

• sensitive,

• specific,

• capable of identifying who will be at risk for peri-operative cardiacevents,

• capable of identifying those patients with correctable coronary lesions.

Risk stratification tests should be restricted to patients clinically assessed as highrisk (and occasionally significant intermediate risk) because:

• limiting their use in this way has been shown to enhance predictiveaccuracy,

• uncontrolled use would have significant resource and cost implications.

A classification of non-invasive risk stratification tests is offered in table 1.8.

Ambulatory electrocardiographic monitoring

This involves making a 24- to 48-h pre-operative ambulatory ECG record. Episodesof ischaemia may be predictive of early and late ischaemic cardiac events.

However, there are a number of limitations:

• It cannot be performed in a significant percentage of patients withbaseline ECG changes as it is less predictive in high-risk groups withabnormal ECGs.


18

Chap-01.qxd 2/1/02 12:03 PM Page 18

• The test only provides a binary outcome (normal/abnormal) and,therefore, cannot further stratify the high-risk group in order to iden-tify the subset for which coronary angiography should be considered.

Ambulatory ECG should, therefore, be restricted to identifying patients for whomadditional surveillance or intervention might be beneficial. It should not be usedas the only diagnostic test to identify patients for coronary angiography.

‘Stress’ testing

• In both animal and human models a coronary stenosis will only start toproduce flow limitation at rest when the cross sectional diameter isreduced by 85–90%.

• Under conditions of maximal flow, however, stenoses with as little as45% reduction in cross sectional diameter will cause flow limitation.

• Clinically this phenomenon manifests as effort-induced angina.

• Thus the sensitivity of diagnostic physiological tests are markedlyimproved with increased flow enabling the detection of lesser stenoses.

Increasing coronary blood flow can be achieved in a number of ways includinggraded exercise, pacing and pharmacological agents, and these methods are com-bined with various methods of detecting ischaemia including ECG, echocardio-graphy and nuclear myocardial imaging.


19

Table 1.8 – Non-invasive tests currently available to stratify cardiovascular risk.

Modality Rest tests ‘Stress’ tests

ECG Resting ECG Ambulatory ECGStress ECG with exerciseStress ECG with inotropes

DobutamineArbutamine

Myocardial nuclear imaging Myocardial rest perfusion imaging Myocardial stress perfusion inCoronary vasodilators

AdenosineDipyridamole

InotropesDobutamineArbutamine

Radionuclide ventriculography

Echocardiography Resting echo Inotrope stress echoDobutamineArbutamine

Invasive/exercise vO2 maxCoronary angiography

Chap-01.qxd 2/1/02 12:03 PM Page 19

Exercise stress testing

Exercise is most commonly employed in exercise ECGs under the Bruce Protocol.

Although not validated for improving outcome, it is extremely useful as a pre-dictor of pre-operative risk as it identifies patients with both poor functionalcapacity and associated ischaemia.19

It is effective at stratifying coronary risk according to:

• degree of functional incapacity,

• ischaemic symptoms,

• ischaemic severity (stage of onset, depth and duration of ST segmentdepression),

• haemodynamic instability,

• electrical instability.

The degree of positivity of test correlates with extent and severity of disease.

The risk of peri-operative cardiac events and long-term risk is significantlyincreased in patients with an abnormal exercise ECG at low workloads.

The sensitivity gradient for detecting obstructive coronary disease is depend-ent on:

• severity of stenosis,

• criteria used for a positive test,

• extent of disease, i.e. the presence of a prior clinical history.

For example, in patients with no cardiac history and a normal resting ECG, only20–25% of patients will have an abnormal exercise ECG, whereas in patients witha prior history of MI or an abnormal rest ECG 35–50% of patients will have anabnormal exercise ECG. However, it should be borne in mind that:

• As many as 50% of patients with significant CAD and adequate levels ofexercise can still produce a normal exercise ECG.35

• There is no evidence to support its use in low-risk groups.

Pharmacological stressors

Inotropes

• Commonly used inotropes include dobutamine and the newer agentarbutamine. These drugs increase myocardial oxygen demand throughinotropic stimulation.


20

Chap-01.qxd 2/1/02 12:03 PM Page 20

• They can often achieve coronary blood flows greater than with exercisebut not as great as with adenosine or dipyridamole.

• They tend to be used in patients with asthma who cannot toleratevasodilators.

• Dobutamine is best avoided in patients with serious arrhythmias andsevere hypertension or hypotension.

Coronary vasodilatorsThese agents produce non-physiological and sometimes supra-physiologicalincreases in coronary flow without altering the metabolic demands of the heart.

• Adenosine is a breakdown product of ATP metabolism and is a potentcoronary vasodilator. In conditions of oxygen starvation rising adeno-sine levels result in coronary vasodilation linking coronary blood flowto myocardial oxygen demand.

• Dipyridamole inhibits the breakdown of free adenosine causing levelsto rise and thereby produces coronary vasodilatation as describedabove.

• Methylxanthines (e.g. caffeine) are competitive inhibitors of adeno-sine at purinergic receptors and patients are advised to avoid these for at least 6–8 h prior to testing achieve maximal coronary vasodilatoreffect.

All these agents can induce or exacerbate bronchospasm, and their use is relativelycontraindicated in asthmatics. Unexpected severe bronchospasm can be readilyantagonised with intravenous aminophylline. Dipyridamole should also be avoidedin patients with critical carotid disease. Other adverse effects include headaches,flushing and hypotension.

Examples of some non-invasive tests that employ pharmacological stressorsinclude:

Dobutamine stress echocardiography

• Provides an opportunity to assess both LV and valvular function.

• Can be performed safely and with acceptable patient tolerance.

• Very accurate in identifying patients with significant angiographiccoronary disease.

• The published experience of dobutamine stress echocardiography toassess peri-operative risk before vascular and non-vascular surgery is rela-tively small compared with the published literature on exercise testingor intravenous dipyridamole myocardial perfusion imaging.


21

Chap-01.qxd 2/1/02 12:03 PM Page 21

• Adding atropine to those patients who fail to meet the target heart rateimproves sensitivity.

• Several studies suggest that the degree of wall motion abnormalitiesand/or wall motion change at low infusion rates of dobutamine is espe-cially important.

Myocardial perfusion stress imaging

• Myocardial uptake of perfusion agents is dependent on blood flowrather than oxygen demand or metabolic activity.

• Resting images alone can be done but sensitivity is very low and is significantly increased if stress images are taken.

• Any of the agents described above can be used to increase coronaryblood flow with the most data available for dipyridamole.

• The tests have high sensitivity and specificity for peri-operative coron-ary events.

The technique involves the injection of a radionuclide during peak blood flow.The main isotopes in use are:

• Thallium-201 is a monovalent cation and like potassium is taken intomyocardial cells by Na–K ATPase. It emits both X-rays and gamma radi-ation and has a relatively long half-life of 73.1h. It readily and rapidlyredistributes with myocardial blood flow so that a 2nd dose is not requiredto obtain rest images. The main disadvantage with thallium is the lowenergy of its emitted radiation resulting in significant soft tissue attenu-ation, which can affect image quality and hinder image interpretation.

• Technetium-99m is an isotope with many advantages over thalliumfor producing better quality myocardial images; it emits radiation at ahigher energy resulting in less attenuation, and its shorter half-life of 6 hallows larger doses. Technetium-99m does not redistribute and, there-fore, requires separate stress and rest injections. The main compoundincorporating this isotope is known as Sestamibi. Unfortunately thesuperior image quality of images produced with Technetium-99m hasyet to be shown to improve diagnostic accuracy.

Images of the myocardium are then compared at peak or stress coronary flow andafter a delay, at rest. Exercise-induced ischaemia results in a perfusion defect on thestress images that fills in or redistributes on the rest images. MI manifests as fixedperfusion defects on both the stress and rest images.

Severe myocardial disease causing exercise-induced pump failure can also bedetected using this technique. The LV decompensates with a fall in stroke volume


22

Chap-01.qxd 2/1/02 12:03 PM Page 22

and ejection fraction with a resultant increase in LV end diastolic pressure anddiameter; on the stress images the ventricular cavity is larger and the ventricle wallis thinner. As a further consequence of the rise in LV end diastolic pressure, raisedlevels of isotope will often accumulate in the lungs.

Subendocardial ischaemia results in reduced endocardial uptake of isotope andmay also appear as LV dilatation.

In increasing grade of risk, possible scan results are:

Normal scan � fixed defects � redistribution defects

Furthermore, as the size of the defect increases, risk significantly increases.

Digital quantitation of scan abnormalities improves positive predictive value and isimproving alongside advances in technology.

The need for caution in routine screening with a dipyridamole–thallium stress testof all patients before vascular surgery was raised by Baron et al.36 In this review of457 patients undergoing elective abdominal aortic surgery, the presence of definiteCAD on clinical assessment and age � 65 years were better predictors of cardiaccomplications than perfusion imaging.

LV ejection fractionResting ventricular function (including LV ejection fraction) is usually deter-mined by echocardiography but may also be determined by radionuclide angio-graphy, gated radionuclide imaging or contrast ventriculography.

Ejection fraction determined by echo is limited in accuracy as the formula used inits calculation assumes the ventricle to be a sphere and the diameters of the ‘fullsphere’ and ‘empty sphere’ are only measured in one plane.

Increased risks of complications are associated with:

• LV ejection fraction � 35%,19

• diastolic and systolic dysfunction are markers for post-operative con-gestive cardiac failure, and in critically ill patients, death.

It is important to note that resting LV function is not a consistent predictor of peri-operative ischaemic events.

Recommendations for non-invasive testing19

In most ambulatory patients, the test of choice is exercise ECG testing, which canboth provide an estimate of functional capacity and detect myocardial ischaemiathrough changes in the ECG and haemodynamic response.


23

Chap-01.qxd 2/1/02 12:03 PM Page 23

In patients with contraindications to exercise ECG testing, e.g. left bundle branchblock or LV hypertrophy, other techniques may be preferable such as:

• exercise echocardiography,

• exercise myocardial perfusion imaging.

In those patients unable to perform adequate exercise the following may be moreuseful:

• dipyridamole–thallium scanning;

• dobutamine echocardiography;

• finally if there is an additional question about valvular dysfunction, theechocardiographic stress test is favoured;

• in many instances, either myocardial stress perfusion imaging or stressechocardiography are appropriate.

In a recent meta-analysis assessing the use of, ambulatory ECG, dobutamine stressechocardiography, radionuclide ventriculography and dipyridamole–thalliumscanning, for predicting adverse cardiac outcome after vascular surgery, all tests hada similar predictive value, with overlapping confidence intervals.

Other important points to note include:

• Local expertise and experience of a test is probably more importantthan the particular type of test.

• Selective rather than routine testing improves cost effectiveness.

• Non-selective blanket tests are not cost effective and are an inefficientuse of resources because at the extremes of risk these tests add very little predictive value.

Invasive testing (coronary angiography)

For very high-risk patients it may be sometimes more appropriate to proceed withcoronary angiography rather than perform a non-invasive test. These might include:

• patients with major clinical predictors,

• advanced ischaemic risk such as unstable angina or residual ischaemiafollowing a recent MI.

In addition it may be reasonable to consider coronary angiography and percu-taneous transluminal coronary angioplasty (PTCA) or coronary artery bypassgrafting (CABG) before non-cardiac surgery when the stress of elective non-cardiacsurgery is likely to exceed the stress of daily life.


24

Chap-01.qxd 2/1/02 12:03 PM Page 24

However, before proceeding to coronary angiography one needs to ascertain thatthe patient is fit enough for either PTCA or CABG. If not angiography only addsto the cost and will not improve outcome.

• As far as the new treatment modality of PTCA is concerned, there is little evidence available at present concerning its pre-operative use tooptimise patients for non-cardiac surgery.

• Indeed there are no controlled trials comparing peri-operative cardiacoutcome after PTCA versus medical therapy.

• A number of small observational series have suggested that cardiacdeath is infrequent in patients who have coronary angioplasty beforenon-cardiac surgery.

• Observational studies have shown a risk reduction for non-cardiac sur-gery after coronary bypass.

PRE-OPERATIVE ASSESSMENT AND RISK

The aim of pre-operative assessment is to minimise morbidity and mortality.

Three questions should be asked when assessing surgical patients with the aim ofminimising operative risk:

1. Is the patient’s medical and physiological status optimum?

2. If not, can the patient’s status be improved (time permitting)?

3. If not, should the operation still proceed? In other words do the risks ofnot operating outweigh the risks of operating. For example, medical status is almost irrelevant if the operation is clearly life saving. Thus,no patient is ‘not fit’ for surgery – it just depends on the urgency of thesituation.

Pre-operative assessment should identify those patients who are at high risk ofpre- or post-operative organ failures. Such patients may need additional moni-toring and may warrant admission to ICU or an HDU post-operatively for organfunction monitoring or support.

Once the risk has been established and attempts have been made to optimise the patient’s condition (i.e. reduce the risk), patients and their families may needto decide whether to proceed with surgery. One of the responsibilities of theanaesthetist is be able to give the patient relevant and accurate information on riskin order to help them decide. Unfortunately few patients will have read the chapter in this text on the meaning of risk!


25

Chap-01.qxd 2/1/02 12:03 PM Page 25

Further reading

Adams AM,Smith AF. Risk perception and communication: recent developmentsand implications for anaesthesia. Anaesthesia 2001; 56: 745–55.

References

1. Rogers FB, Simons R, Hoyt DB et al. In-house board-certified surgeonsimprove outcome for severely injured patients: a comparison of two universitycenters. J Trauma 1993; 34: 871–5.

2. Reynolds HN, Haupt MT, Thill-Baharozian MC et al. Impact of critical carephysician staffing on patients with septic shock in a university hospital medicalintensive care unit. JAMA 1988; 260: 3446–50.

3. Harmon JW, Tang DG, Gordon TA et al. Hospital volume can serve as a sur-rogate for surgeon volume for achieving excellent outcomes in colorectalresection. Ann Surg 1999; 230: 404–11.

4. Merry AF,Ramage MC,Whitlock RM et al. First-time coronary artery bypassgrafting: the anaesthetist as a risk factor. Br J Anaesth 1992; 68: 6–12.

5. Campling EA, Devlin HB, Hoile RW et al. Who operates when. The report of the National Confidential Enquiry into Perioperative Deaths 1996/1997.NCEPOD, London, 1997.

6. Stidham KR, Johnson JL, Seigler HF. Survival superiority of females withmelanoma. A multivariate analysis of 6383 patients exploring the significanceof gender in prognostic outcome. Arch Surg 1994; 129: 316–24.

7. Wichmann MW, Inthorn D, Andress HJ et al. Incidence and mortality ofsevere sepsis in surgical intensive care patients: the influence of patient genderon disease process and outcome. Int Care Med 2000; 26: 167–72.

8. Offner PJ, Moore EE, Biffl WL. Male gender is a risk factor for major infec-tions after surgery. Arch Surg 1999; 134: 935–8.

9. Kollef MH, O’Brien JD, Silver P. The impact of gender on outcome frommechanical ventilation. Chest 1997; 111: 434–41.

10. Norman PE, Semmens JB, Lawrence-Brown M et al. The influence of genderon outcome following peripheral vascular surgery: a review. Cardiovasc Surg2000; 8: 111–15.

11. Moul JW, Douglas TH, McCarthy WF et al. Black race is an adverse prognos-tic factor for prostate cancer recurrence following radical prostatectomy in anequal access health care setting. J Urol 1996; 155: 1667–73.


26

Chap-01.qxd 2/1/02 12:03 PM Page 26

12. Connell PP, Rotmensch J, Waggoner SE et al. Race and clinical outcome inendometrial carcinoma. Obstet Gynecol 1999; 94: 713–20.

13. Andersson B, Sylven C. The DD genotype of the angiotensin-convertingenzyme gene is associated with increased mortality in idiopathic heart failure.J Am Coll Cardiol 1996; 28: 162–7.

14. Poeze M,Takala J,Greve JWM,Ramsay G. Pre-operative tonometry is predict-ive for mortality and morbidity in high-risk surgical patients. Int Care Med2000; 26: 1272–81.

15. Kelion AD, Banning AP. Is simple clinical assessment adequate for cardiac riskstratification before elective non-cardiac surgery. Lancet 1999; 354: 1838.

16. Castelli WP. Epidemiology of coronary heart disease: the Framingham study.Am J Med 1984; 76: 4–12.

17. Becker RC, Terrin M, Ross R et al. and the Thrombolysis in MyocardialInfarction Investigators. Comparison of clinical outcomes for women andmen after acute myocardial infarction. Ann Intern Med 1994; 120: 638–45.

18. Shah KB, Kleinman BS, Rao TLK et al. Angina and other risk factors inpatients with cardiac diseases undergoing non-cardiac operations. AnesthAnalg 1990; 70: 240–7.

19. Eagle KA, Brundage BH, Chaitman BR et al. Guidelines for perioperative cardiovascular evaluation for non-cardiac surgery: an abridged version of thereport of the American College of Cardiology/American Heart AssociationTask Force on Practice Guidelines. J Am Coll Cardiol 1996; 27: 910–48.

20. Goldman L, Caldera DL. Risks of general anesthesia and elective operation inthe hypertensive patient. Anesthesiology 1979; 50: 285–92.

21. Goldman L, Caldera DL, Nussbaum SR et al. Multifactorial index of cardiacrisk in non-cardiac surgical procedures. N Engl J Med 1977; 297: 845–50.

22. Reyes VP, Raju BS, Wynne J, Stephenson et al. Percutaneous balloon valvulo-plasty compared with open surgical commissurotomy for mitral stenosis.N Engl J Med 1994; 331: 961–96.

23. Alpert JS, Chipkin SR, Aronin N. Diabetes mellitus and silent myocardialischemia. Adv Cardiol 1990; 37: 279–303.

24. ASA. New classification of physical status. Anaesthesiology 1963; 24: 111.

25. Detsky AS, Abrams HB, Forbath N, Scott JG, Hillard JR. Cardiac assessmentfor patients undergoing non-cardiac surgery. A multifactorial clinical riskindex. Arch Int Med 1986; 146: 2131–4.


27

Chap-01.qxd 2/1/02 12:03 PM Page 27

26. Lee TH, Marcantonio ER, Mangione CM et al. Derivation and prospectivevalidation of a simple index for prediction of cardiac risk of major non-cardiacsurgery. Circulation 1999; 100: 1043–9.

27. Wolters U, Wolf T, Stutzer H, Schroder T. ASA classification and peri-operative variables as predictors of postoperative outcome. Br J Anaes 1996;77: 217–22.

28. Copeland GP, Jones D, Waiters M. POSSUM: a scoring system for surgicalaudit. Br J Surg 1991; 78: 355–60.

29. Lazarides MK,Arvanitis DP, Drista H et al. POSSUM and APACHE II scoresdo not predict the outcome of ruptured infrarenal aortic aneurysms. Ann VascSurg 1997; 11: 155–8.

30. Jones DR, Copeland GP, de Cossart L. Comparison of POSSUM withAPACHE II for prediction of outcome from a surgical high-dependency unit.Br J Surg 1992; 79: 1293–6.

31. Sagar PM, Hartley MN, Mancey-Jones B et al. Comparative audit of colorec-tal resection with the POSSUM scoring system. Br J Surg 1994; 81: 1492–4.

32. Hlatky MA, Boineau RE, Higginbotham MB, Lee KL, Mark DB, Califf RM,Cobb FR, Pryor DB. A brief self-administered questionnaire to determinefunctional capacity (the Duke Activity Status Index). Am J Cardiol 1989; 64:651–4.

33. Mangano DT. Perioperative cardiac morbidity. Anesthesiology 1990; 72:153–84.

34. Taylor LM Jr, Porter JM. Basic data related to clinical decision-making inabdominal aortic aneurysms. Ann Vasc Surg 1987; 1: 502–4.

35. Chaitman BR. The changing role of the exercise electrocardiogram as a diag-nostic and prognostic test for chronic ischemic heart disease. J Am Coll Cardiol1986; 8: 1195–210.

36. Baron JF, Mundler O, Bertrand M, Vicaut E, Barre E, Godet G, Samama CM et al. Dipyridamole–thallium scintigraphy and gated radionuclide angiographyto assess cardiac risk before abdominal aortic surgery. N Engl J Med 1994; 330:663–9.


28

Chap-01.qxd 2/1/02 12:03 PM Page 28

29

2RESPIRATORY RISK AND

COMPLICATIONS

There is now an increasing volume of literature regarding the identification of thepatient at risk of respiratory complications:

• Respiratory complications are at least as, and sometimes more commonthan cardiac complications.1

However, reaching a consensus on what constitutes a postoperative respiratorycomplication has proved difficult in recent years and has significantly hinderedresearch in this area. Improvements in the health of the population, advances inanaesthesia and surgery and reduction in prevalence of smoking amongst the popu-lation have all combined to cause problems interpreting the significance of someof the early studies on respiratory risk and complications.

It is now increasingly gaining acceptance that significant respiratory complicationsare those that affect outcome. These are problems that prolong hospital or inten-sive care unit (ICU) stay, or alternatively contribute to morbidity or mortality.2–4

They include:

• pneumonia,

• respiratory failure requiring mechanical ventilation,

• bronchospasm,

• atelectasis,

• exacerbation of chronic underlying disease.

Factors that have been shown to be predictive of postoperative pulmonary respiratory complications can be subdivided into patient- and surgery-related risk factors.

Chap-02.qxd 2/2/02 12:54 PM Page 29

PATIENT-RELATED FACTORS

The relative increases in risk associated with the presence of patient-related factorsis given in table 2.1.

Smoking

Smoking has long been recognised as a respiratory risk factor for patients bothwith and without chronic lung disease, and causes a 3–4-fold increase in the inci-dence of pulmonary complications. Problems arise from

• increase sputum and mucus production,

• impaired ciliary clearance of secretions in the lung,

• increases in airway reactivity,

• reduced oxygen delivery to the tissues from increased carboxyhaemo-globin levels,

• harmful effects of nicotine on heart including tachycardia and vaso-constriction.

The increased risk from smoking declines after 8 weeks abstinence.5

Interestingly, smokers who have abstained for � 8 weeks are at increased risk ofpostoperative respiratory complications.

General health

• Both the Goldman Cardiac Risk Index and American Society ofAnaesthesiologists (ASA) status (as indicators of poor general health) arepredictive for pulmonary as well as cardiac complications.6,7 These arediscussed in Chapter 1.


30

Table 2.1 – Factors causing a relative increase in risk of pulmonary complications in unselected surgery.

Risk factor Unadjusted relative risk associated with factor

Smoking 3.4Poor general health ASA � II 1.7Age � 70 1.9–2.4Obesity 1.3COPD 2.7–3.6

Adapted from Ref. 9.

Chap-02.qxd 2/2/02 12:54 PM Page 30

• Indeed, postoperative respiratory compromise often accompanies cardiacdysfunction and vice versa, and is a reflection of the complex relation-ship between cardiac and respiratory pathophysiology.

Exercise capacity

• Poor exercise tolerance is strongly predictive for respiratory com-plications.7,8

Age

• Age has not been identified as independent variable predictive ofincreased pulmonary risk.

• Respiratory complications are more related to coexisting conditions(more common in the elderly) than directly to age.

The changes in the respiratory system and their anaesthetic implications are dis-cussed in detail in Chapter 7.

Obesity

• Morbid obesity may predispose to ventilatory problems in the immediatepostoperative period.

• However, contrary to popular belief, obesity has not been proven to bea significant risk factor predisposing to significant respiratory complica-tions as defined above.9,10

• In a recent review, it was shown that obesity posed only a relativeincreased risk of 1.3-fold for respiratory complications (table 2.1).

One should note, however, that morbid obesity is a significant general risk factorin the surgical patient. The main problems in the morbidly obese are

• increased incidence of hypertension and ischaemic heart disease;

• mechanical respiratory problems i.e. decreased compliance and func-tional residual capacity, especially when supine;

• increased incidence of perioperative infection including wound andrespiratory infections;

• increased incidence of abdominal wound dehiscense;

• increased rate of DVT;

• increased oesophageal reflux leading to increased risk of perioperativegastric aspiration.

RESPIRATORY RISK AND COMPLICATIONS

31

Chap-02.qxd 2/2/02 12:54 PM Page 31

Chronic obstructive pulmonary disease

• Patients with chronic obstructive pulmonary disease (COPD) are atincreased risk of postoperative respiratory complications, the level ofincreased risk related to the severity of the lung disease.9 Hypercapniais particulary ominous.

• Patients with COPD should be optimised prior to surgery with theusual therapies and those with acute exacerbations should be deferreduntil treated.

Asthma

• Studies conflict over whether asthma increases the risk of respiratorycomplications as defined above.

• The emergency patient presenting with severe, symptomatic bron-chospasm, however, is a serious challenge necessitating aggressive treat-ment with bronchodilators. Airway instrumentation in these patientsmay precipitate intractable bronchospasm.

SURGERY-RELATED FACTORS

Anatomical site of surgery

Even with the patient-related factors described above taken into account,the anatomical site of surgery remains the most important predictor of respiratory risk:9

• The risk of pulmonary complications is directly related to the proximityof the incision to the diaphragm (table 2.2).


Table 2.2 – Effect of surgical site on postoperative pulmonary complications.

Anatomic site of surgery Mean incidence of pulmonary complications (%)

Thoracic 23Upper abdominal 22Lower abdominal 9Other 5Laparoscopic 0.3–0.4


32

Chap-02.qxd 2/2/02 12:54 PM Page 32

• Respiratory complications are rare if surgery is outside the thorax orabdomen.9

Duration of surgery

• Surgery longer than 3 h duration increases the likelihood of respiratorycomplications by 1.6–5.2-fold.11,12

• Thus, in those patients with significant predictors of respiratory risk,one should aim for as short a procedure as possible, ideally performedby the most efficient surgeon.

General anaesthesia and muscle relaxants

• The use of general as opposed to epidural or spinal anaesthesia multi-plies the respiratory risk by 1.2 to 3 times (table 2.3).9

• The use of muscle relaxant pancuronium in one study resulted in morethan a 3-fold increase in respiratory complications when compared tothe use of atracurium or vecuronium.13 Complications probably arosethrough inadequate reversal, causing hypoventilation and a reducedability to cough. The authors, therefore, recommended that pancuro-nium should be avoided in high-risk respiratory patients.

Preoperative assessment

This includes history, clinical examination, and relevant investigations:

• Patients should be asked about dyspnoea, productive or chronic cough,and exercise tolerance.

• Physical findings on examination, predictive of complications, includeall those signs of acute and chronic lung disease including reduced airentry, percussive dullness, and wheezes or crackles.


33

Table 2.3 – Other surgical factors affecting pulmonary risk.

Risk factor Site of surgery Unadjusted relative risk associated with factor

Surgery � 3 h duration Unselected 1.6–5.2Thoracic or abdominal 3.6

General anaesthesia Unselected 1.2–�Thoracic or abdominal or vascular 2.2–3.0

Use of muscle relaxants Unselected 3.2


Chap-02.qxd 2/2/02 12:54 PM Page 33

PULMONARY FUNCTION TESTING

The role of pulmonary function testing (PFT) is controversial. It is generally agreedthat PFTs

• can detect the presence or absence of pulmonary disease;

• can determine the types of defect present; i.e. restrictive, obstructive, or diffusion defect;

• can grade the extent or severity of the defect;

• are highly reproducible;14

• correlate with all cause mortality.14

Historically, early studies on PFT were based on patients undergoing thoracic surgery for diseases including lung cancer and tuberculosis, and cut-off thresholdswere established below which surgery was deemed contraindicated. Over timethese values have subsequently been applied to non-thoracic surgical populations.

It is now gaining acceptance that low values only indicate the severity of currentdisease, but do not reliably predict complications or outcome.9

For example, one study15 attempted to determine if severe airflow obstructionreliably predicted the incidence of postoperative complications, by comparingpatients undergoing abdominal surgery who had significant airflow obstructionwith forced expiratory volume 1 (FEV1) � 40% with a group of controls. Theyfound that airflow obstruction only predicted postoperative bronchospasm.Outcome parameters were not different between the two groups:

• The important question is whether PFT adds to the informationalready gained from clinical assessment. Many authors do not believethey do, as preoperative history and respiratory examination is moresensitive and specific than abnormal preoperative spirometry.8

• In addition, in studies that have compared both clinical assessment andPFTs, clinical findings are more predictive of respiratory complicationsthan are PFTs.8,16,17

• A review on this subject in 198918 concluded that there were numerousmethodological limitations in existing studies that prevented them fromconcluding that PFTs helped to predict those at risk of pulmonary com-plications.

• More recently, PFTs have been shown to have inconsistent and variablepredictive value.9

There is no evidence that PFTs improve the ability to identify patients at risk whohave no clinical findings suggestive of pulmonary disease on history or examination.


34

Chap-02.qxd 2/2/02 12:54 PM Page 34

However, it must be emphasised that although not definitely predictive of respira-tory complications, PFTs do provide an indication of disease severity and this mayprompt further optimisation and direction into risk reduction strategies, whichmay contribute to an improved outcome.

PFTs to determine suitability for surgery

Some advocate PFTs as a tool to identify those patients at such high risk that sur-gery should be avoided (see table 2.4 for commonly quoted thresholds).

This philosophy is increasingly being refuted for the following reasons:

• The accuracy and meaning of predicted postoperative spirometry values have been called into question (see above).

• Patients with PFTs suggestive of extreme risk are increasingly beingshown to have an acceptable rate of pulmonary complications. This isbecause recent developments in risk reduction strategies have improvedoutcomes.

• Lung volume reduction surgery as a treatment for end stage emphysemain patients with spirometry values that hitherto were below previouslydefined cut-off thresholds is increasingly accepted (see below).

• One study2 studied 107 patients undergoing surgery with anFEV1 � 50% and FEV1/forced vital capacity (FVC) ratio of �70%;postoperative respiratory complications occurred in only 29% of thepatients, and there was only one death amongst 97 which underwentnon-cardiac procedures.

There is still, however, a general consensus that all patients undergoing lung resec-tion surgery should have PFTs performed. The reasons often cited include:

• To provide a baseline with which to compare postoperative values.


35

Table 2.4 – Preoperative spirometry with previously reported thresholds forincreased pulmonary risk.

True volume Percentage of predicted volume achieved

FVC% High risk if � 70%8

FEV1% High risk if � 70%8

FEV1/FVC% High risk if � 65%25

Predicted postoperative volume Percentage of predicted postoperative volumeppoFEV1 High risk � 800 ml ppoFEV1% High risk � 40%19

Percentage predicted volumes achieved based on population normals for age, sex and height.Adapted from Refs 9, 19, 26.

Chap-02.qxd 2/2/02 12:54 PM Page 35

• To attempt to predict who will not tolerate lung resection. Predictedpostoperative values are derived from an estimation of the number offunctional lung units expected to remain after surgery.

• The most widely accepted threshold for lung resection surgery has beena predicted postoperative FEV1 (ppoFEV1) � 800 ml19 (table 2.4).

• The accuracy of these predictive models have been questioned recentlyin studies that have compared predicted with actual postoperative val-ues20,21 and showed that the predictions tended to be overly pessimisticand overestimated the loss of functional lung units. In addition, aspatients recovered in the months following surgery, their spirometryresults progressively improved.

Lung volume reduction surgery – confounding the PFTsIn lung volume reduction surgery, bilateral wedges of hyperinflated lung areresected to improve elastic recoil pressure and diaphragm position and function.The procedure is rapidly gaining acceptance as therapy for patients with

• end stage emphysema, including even those so severe as to be listed forlung transplantation,

• severe emphysema and malignancy (which previously would have beenassessed as inoperable on the basis of spirometry results).

These patients typically have FEV1 � 35% or � 0.5 l and may also be ventilatordependant. In one study,FEV1 ranged from 0.23 to 0.5 l. Here dyspnoea improvedin 89% of patients and mean FEV1 improved by 51% and FVC increased by 56%.22

Thus, there is now increasing evidence that PFTs should not be used to denypatients’ surgery,which in the case of lung malignancy may potentially be curative.9

How should we select which patients should undergo PFT?The American College of Physicians recommendations published in 199022 andthe recent review by Smetana9 suggest that PFT be limited to the followingpatient groups:

• All patients undergoing lung resection surgery.

• Patients undergoing thoracic, cardiac, or upper abdominal surgery who also either have a history of smoking or symptoms of cough, dyspnoea,or unaccountable exercise intolerance.

• Patients undergoing head and neck, orthopaedic or lower abdominal surgery with unexplained dyspnoea or pulmonary symptoms.

• Patients with COPD or asthma to determine if their airflow obstructionis optimised.


36

Chap-02.qxd 2/2/02 12:54 PM Page 36

ARTERIAL BLOOD GAS ANALYSIS

The American College of Physicians recommends blood gas analysis in patientswith a history of dyspnoea and tobacco use who are undergoing upper abdomi-nal or coronary artery surgery.23

Hypercapnia

A number of small studies have suggested that a pCO2 � 45 predisposes to pul-monary complications.24 However, all those with high pCO2 also had substantialairflow obstruction on spirometry and could be more identified more easily bythis less invasive method.

In patients undergoing lung volume reduction surgery or lung resection, hyper-capnia is not predictive for postoperative pulmonary complications.

Arterial pO2

In the study by Nunn25 in 1988, dyspnoea at rest was more predictive than a lowpO2 for identifying those at increased risk of postoperative ventilation.

Further reading

Johnson BD, Beck KC, Zeballos RJ. Advances in pulmonary laboratory testing.Chest 1999; 116: 1377–87.

Ferguson MK. Preoperative assessment of pulmonary risk. Chest 1999; 115(5 suppl.): 58S–63S.

Doyle RL. Assessing and modifying the risk of postoperative pulmonary compli-cations. Chest 1999; 115 (5 suppl.): 77S–81S.

References

1. Lawrence VA, Hilsenbeck SG, Mulrow CD, Dhanda R, Sapp J, Page CP.Incidence and hospital stay for cardiac and pulmonary complications afterabdominal surgery. J Gen Intern Med 1995; 10: 671–8.

2. Kroenke K, Lawrence VA, Theroux JF, Tuley MR. Operative risk in patientswith severe obstructive pulmonary disease. Arch Intern Med 1992; 152: 967–71.

3. Pedersen T,Eliasen K,Henriksen E. A prospective study of risk factors and car-diopulmonary complications associated with anaesthesia and surgery: risk indi-cators of cardiopulmonary morbidity. Acta Anaesthesiol Scand 1990; 34: 144–55.

4. Gracey DR, Divertie MB, Didier EP. Preoperative pulmonary preparation ofpatients with chronic obstructive pulmonary disease: a prospective study.Chest 1979; 76: 123–9.


37

Chap-02.qxd 2/2/02 12:54 PM Page 37

5. Warner MA, Offord KP, Warner ME, Lennon RL, Conover MA, Jansson-Schumacher U. Role of preoperative cessation of smoking and other factorsin postoperative pulmonary complications: a blinded prospective study ofcoronary artery bypass patients. Mayo Clin Proc 1989; 64: 609–16.

6. Lawrence VA, Dhanda R, Hilsenbeck SG, Page CP. Risk of pulmonary complications after elective abdominal surgery. Chest 1996; 110: 744–50.

7. Gerson MC, Hurst JM, Hertzberg VS, Baughman R, Rouan GW, Ellis K.Prediction of cardiac and pulmonary complications related to elective abdom-inal and non-cardiac thoracic surgery in geriatric patients. Am J Med 1990; 88:101–7.

8. Williams-Russo P, Charlson ME, MacKenzie CR, Gold JP, Shires GT.Predicting postoperative pulmonary complications: is it a real problem? ArchIntern Med 1992; 152: 1209–13.

9. Smetana GW. Preoperative pulmonary evaluation. N Engl J Med 1999; 340:937–44.

10. Phillips EH, Carroll BJ, Fallas MJ, Pearlstein AR. Comparison of laparoscopiccholecystectomy in obese and non-obese patients. Am Surg 1994; 60: 316–21.

11. Brooks-Brunn JA. Predictors of postoperative pulmonary complications following abdominal surgery. Chest 1997; 111: 564–71.

12. Celli BR, Rodriguez KS, Snider GL. A controlled trial of intermittent posi-tive pressure breathing, incentive spirometry, and deep breathing exercises inpreventing pulmonary complications after abdominal surgery. Am Rev RespirDis 1984; 130: 12–15.

13. Berg H, Viby-Mogensen J, Roed J et al. Residual neuromuscular block is a risk factor for postoperative pulmonary complications: a prospective, ran-domised, and blinded study of postoperative pulmonary complications afteratracurium, vecuronium, and pancuronium. Acta Anaesthesiol Scand 1997; 41:1095–103.

14. Kimball WR. The role of spirometry in predicting pulmonary complicationsafter abdominal surgery progressing toward an answer. Anesthesiology 1999; 90:353–9.

15. Warner DO, Warner MA, Offord KP, Schroeder DR, Maxson P, Scanlon PD.Airway obstruction and perioperative complications in smokers undergoingabdominal surgery. Anesthesiology 1999; 90: 372–9.

16. Kroenke K, Lawrence VA, Theroux JF, Tuley MR, Hilsenbeck S. Post-operative complications after thoracic and major abdominal surgery in patientswith and without obstructive lung disease. Chest 1993; 104: 1445–51.


38

Chap-02.qxd 2/2/02 12:54 PM Page 38

17. Cain HD, Stevens PM, Adaniya R. Preoperative pulmonary function andcomplications after cardiovascular surgery. Chest 1979; 76: 130–5.

18. Lawrence VA, Page CP, Harris GD. Preoperative spirometry before abdominaloperations: a critical appraisal of its predictive value. Arch Intern Med 1989;149: 280–5.

19. Olsen G,Block A, Swenson E et al. Pulmonary function evaluation of the lungresection candidate: a prospective study. Am Rev Respir Dis 1975; 111:379–87.

20. Bolliger C, Perruchoud A. Functional evaluation of the lung resection candi-date. Eur Respir J 1998; 11: 198–212.

21. Larsen K, Svendsen U, Milman N et al. Cardiopulmonary function at rest andduring exercise after resection for bronchial carcinoma. Ann Thorac Surg 1997;64: 960–4.

22. Eugene J, Dajee A, Kayaleh R et al. Reduction pneumoplasty for patients witha forced expiratory volume in 1 second of 500 millilitres or less. Ann ThoracSurg 1997; 96: 894–900.

23. American College of Physicians. Preoperative pulmonary function testing.Ann Intern Med 1990; 112: 793–4.

24. Milledge JS, Nunn JF. Criteria of fitness for anaesthesia in patients withchronic obstructive lung disease. Br Med J 1975; 3: 670–3.

25. Nunn JF, Milledge JS, Chen D, Dore C. Respiratory criteria of fitness for surgery and anaesthesia. Anaesthesia 1988; 43: 543–51.

26. Gass G, Olsen G. Preoperative pulmonary function testing to predict post-operative morbidity and mortality. Chest 1986; 89: 127–35.


39

Chap-02.qxd 2/2/02 12:54 PM Page 39

Chap-02.qxd 2/2/02 12:54 PM Page 40


41

3LESSONS FROM THE NATIONAL

CONFIDENTIAL ENQUIRY INTOPERIOPERATIVE DEATHS

The National Confidential Enquiry into Perioperative Deaths (NCEPOD) hasproduced its 10th report.1 They take the opportunity to reflect on their own con-tribution to improving the quality of patient care since publication of their firstreport in June 1990.2

It is also a convenient time to reflect on the issues highlighted by NCEPOD andthe lessons to be learnt in the management of the high risk surgical patient.

Note: The NCEPOD considers the quality of the delivery of care and not speci-fically causation of death.

• Data is collected from all NHS and Defence Secondary Care Agencyhospitals in England, Wales and Northern Ireland, and public hospitalsin Guernsey, Jersey and the Isle of Man, as well as many hospitals in theindependent sector.

• NCEPOD collects basic details on all deaths in hospital within 30 daysof a surgical procedure, through a system of local reporting.

• Since the introduction of clinical governance in April 1999, participationin the confidential enquiries has become a mandatory requirement forclinicians in the NHS.

• The data collection runs from 1st April to 31st March.

• Each year a sample is selected for more detailed review.

• Questionnaires were returned by 83% surgeons and 85% of anaesthetistsfor the 1998/99 report.3

Chap-03.qxd 2/1/02 12:04 PM Page 41

• The NCEPOD clinical coordinators, together with the advisory groupsfor anaesthesia and surgery, review the completed questionnaires andthen aggregate the data to produce a final report.

NCEPOD does not attempt to collect denominator data or calculate mortalityfigures. Data submitted to the Department of Health as hospital episode statisticsis used to calculate NHS Performance Indicators.

The Performance Indicators for 1998/993 reveal:

• 32 956 deaths in hospital within 30 days of an operative procedure.

• 24 920 after emergency surgery and 8036 after non-emergency surgery.

• A total of 2.3 million procedures* were undertaken (of which 26% wereemergencies).

• Mortality rate of 1.4% after emergency surgery.

• Mortality rate of 0.5% after non-emergency surgery.

NCEPOD does tell us something about the distribution of these deaths:1

• the vast majority of patients are elderly; 70% of deaths occurred inpatients � 70 years;

• 94% of patients have a coexisting medical disease;

• death occurs within 5 days of an operation in almost half of the patientsreported.

Most of the recommendations taken from the 10 reports and presented below represent nothing more than common sense and good clinical practice. Many ofthe recommendations have been repeated over and over again.

NCEPOD RECOMMENDATIONS

Facilities

Individual clinicians efforts to provide the level of care they know is required forthe high risk surgical patient is often frustratingly thwarted by lack of facilities.NCEPOD has helped identify these shortcomings. One of the major lessons to be learnt is that to provide the highest quality of care for these patients, acute surgical services may need to be concentrated on fewer well staffed and resourcedhospitals.4


42

* Definition of procedures used for Performance Indicators is not directly comparable to the definitionsused by NCEPOD.

Chap-03.qxd 2/1/02 12:04 PM Page 42

• A dedicated emergency theatre and recovery should be staffed andavailable 24 h a day. The aim should be to deal with emergency casesduring the working day and avoid out of hours operating.1,5–7

• There should be easy access to a high dependency unit (HDU) andintensive care unit (ICU) on a single site.2,7,8

• An orthopaedic trauma theatre operating during the day with seniorstaff.9

• Elderly patients should not have to wait more than 24 h (once fit) foroperation.10 When a decision to operate is taken there should be acommitment by the clinicians and adequate facilities available to pro-vide appropriate critical care post-operatively.10

• Local protocols should be in place to ensure immediate access to bloodand blood products.11

• CT scanning and availability of neurosurgical consultation should beavailable in any hospital receiving trauma patients.9

• A fibreoptic laryngoscope should be available with trained and nomi-nated staff able to use it.12

• Children’s services should be concentrated to avoid occasional practice.Local arrangements should be in place for the skilled transport of criti-cally ill children when appropriate.1

• An arbitrator/coordinator should exist to ensure emergency cases areprioritised appropriately and emergency theatre space is utilised efficiently.7

Personnel

It is clear that the high risk surgical patient should be directly cared for by experi-enced senior anaesthetists and surgical members of staff. This recommendationhas been one of the cornerstones of the NCEPOD reports over the years:7–10

• In the most recent NCEPOD report1 of the cases sampled 59% of anaes-thetics were given by a consultant and 52% of operations were performedby a consultant surgeon.

• Consultant presence in these cases has changed little since the first report.2

• Comparing NCEPOD 20001 to NCEPOD 19902 fewer of the sampledpatients are anaesthetised or operated on by junior trainees. Howeverthere is now a trend towards a greater reliance on non-consultant careergrade (NCCG) anaesthetists and surgical members of staff.

LESSONS FROM THE NCEPOD

43

Chap-03.qxd 2/1/02 12:04 PM Page 43

It may on occasions be appropriate or unavoidable that consultants cannot bedirectly involved in patient care. At the very least the consultant has a vital role inproviding advice, support and making crucial decisions:5,9

• NCEPOD 20001 reported that only 5% of cases had no consultant surgeon involvement.

• Disappointingly anaesthetists sought advice from a colleague who wasnot present during the anaesthetic in 15% of cases.1

There are a number of reasons why senior help may not be requested for the highrisk patient:

• Insufficient experience to identify the at risk patient.

• Inability to recognise personal limitations.

• Lack of familiarity with local procedures.

• Practical barriers to communication.

• Poor understanding of personal limits of responsibility.

Though these are mainly personal shortcomings senior clinicians should ensurethat the system within their hospital is robust enough to ensure these individuallimitations do not compromise patient safety and good practice.

NCEPOD does provide some practical lessons to enable us to achieve the recom-mendation8 of matching senior surgical and anaesthetic skills to the condition ofthe patient.

Supervision

• Ensure local guidelines are in place, so trainees are clear when to ask forhelp.8,9

• National or regional guidelines may be preferable to avoid confusionwhen trainees rotate between hospitals.8

• Guidelines are particularly important in paediatrics, to prevent occasionalpractice, ensure practitioners retain skills and to promptly identify childrenthat require specialist care and transfer.1,2

Communication

• All staff (including consultants) must be aware of their limitations andwork in an environment where they are encouraged to and are comfort-able with asking for help.5

• Anaesthetists should be consulted as opposed to informed about cases.1


44

Chap-03.qxd 2/1/02 12:04 PM Page 44

• Encourage a team approach between surgeons,anaesthetists and physiciansfor complex cases.6

• Ensure there is adequate communication between specialties and betweengrades of staff within a specialty.11

• ‘Who to call’, ‘When to call’ and ‘How to call’ for help should be easily available and clearly understood.1,7

Staff availability

• All staff covering emergencies should be free from other commitmentsand easily available.

• Emergency theatre sessions should be staffed by consultants.1,7,8

Locums and NCCG anaesthetists and surgeons

• Ensure NCCG doctors within the hospital are aware of their role andlimit of responsibility. They should have equal access to supervision,involvement in audit and opportunities for continued professionaldevelopment.1

• Supervising consultants should be aware of the abilities of locum doctorsbefore appointment.5

• Extra effort and vigilance is required to ensure that locums are appraisedof local guidelines and afforded the same degree of supervision and support as other members of staff.5,11

The Royal College of Anaesthetists provides guidance on the level of supervisionof doctors in training.13 It highlights the statement from the Clinical NegligenceScheme for Trusts (CNST) which increases the statutory requirement to ensurethat doctors taking up post in a new hospital are adequately trained and compe-tent to fulfil the role for which they have been employed. There is now a require-ment for supervisors to list technical skills new doctors are expected to performand, in turn, for the new doctors to indicate their competence to perform the specific tasks. A supervised training programme must rectify any deficiencies ininitial, or continuing competence.

Pre-operative assessment and preparation

In the 10 years that NCEPOD has reported the profile of patients dying within 30 days of their operation has changed.1

Patients are more likely to be older, have undergone an urgent operation, be of poorer physicalstatus and have a coexisting cardiovascular or neurological disorder.


45

Chap-03.qxd 2/1/02 12:04 PM Page 45

Adequate pre-operative assessment and preparation for theatre is vital in the management of the high risk surgical patient.

The management of the patient at this time will greatly influence the subsequentoutcome of the operation and it is crucial that appropriate care is given and appro-priate decisions made to prevent avoidable morbidity and mortality.

• The decision to operate and when not to operate should be made byconsultants.2,12 It is important that there is direct communicationbetween consultant surgeon and anaesthetist and decisions are madejointly.1

• NCEPOD has frequently referred to the problem of operations beingperformed on moribund patients or where the objective of surgery isunclear.2,12 These patients clearly need a consultant evaluation and asensitive but honest approach to the patient and relatives. Too oftenhasty decisions to opt for surgery are made without fully consideringtheir ramifications.

• If the decision to operate or not is contra to that of the patient or rela-tives, having a mechanism to allow discussion with consultant colleaguesfrom the same or allied specialty can be invaluable in reassuring the patientand relatives that an appropriate decision has been made.

• NCEPOD has highlighted the need to provide junior members of staffwith guidance on when to ask for help.7 NCEPOD has indicated theinadequacies and inconsistencies of the ASA classification if used as anassessment tool.1 The ASA classification may not on its own be an ade-quate trigger for alerting inexperienced doctors to a high risk patient.NCEPOD suggested that the use of the P-POSSUM score should bemore widespread.10

• The importance of avoiding rushing patients to theatre before adequateresuscitation is a recurring theme of NCEPOD.2,8 Emergency patientsinvariably require appropriate fluid resuscitation prior to theatre. Insome patients this can safely be undertaken on the ward. In otherpatients, particularly the elderly, this may require a critical care environ-ment and appropriate invasive monitors.1,7

• Poor understanding of fluid management especially in the elderly isoften cited as a contributing factor in NCEPOD cases.6–8,10 The use ofpre-operative critical care services for resuscitation and pre-operativepreparation may well avoid the need for lengthy post-operative criticalcare.7

• Finding a critical care bed pre-operatively in many hospitals may beimpossible. The ability to utilise theatre recovery beds prior to surgery


46

Chap-03.qxd 2/1/02 12:04 PM Page 46

and the development of critical care outreach services may be a short-term solution.

• Whatever the particular local solution it is important to have a mechanismin place to allow patients to be adequately resuscitated in an appropriateenvironment by knowledgeable staff.

• Starting a high risk case without first identifying adequate critical carefacilities post-operatively is to be avoided.6 Consultation with colleagueswho control these beds at the earliest opportunity is essential. It is notalways easy to identify those patients who require HDU care. CEPODhas called for simple nationally agreed criteria to help assess the need forHDU care.

• Over the 10 years of NCEPOD the percentage of patients with coexist-ing medical disorders has increased from 89% to 94%.1 Cardiac disordershave increased from 54% to 66%. NCEPOD suggest that Echocardio-graphy should be available and used more widely in pre-operativeassessments.1 For complex medical disorders the advice of a specialistphysician may be invaluable. NCEPOD would like to see hospitalsdevelop an organisational structure to allow prompt medical reviewshould it be required.1

• Thromboembolic complications continue to be a major cause of mor-bidity and mortality. CEPOD has recognised this in all its reports andhighlighted the inconsistent nature of prophylactic measures. It recom-mends the development of guidelines and clear definition of responsi-bility for implementing prophylactic measures. The guidelines need tobe audited regularly to ensure compliance and efficacy.1,7,8

• Individuals dealing with high risk patients in the pre-operative periodshould be aware of the importance of thromboembolis prophylaxis.

Audit

CEPOD recognises that audit can be a useful tool locally to help improve themanagement of high risk surgery. There is a lack of consistency in the participationin audit both between hospitals and within surgical specialties and anaesthesia.

Of cases sampled for NCEPOD 20001 1/3 of deaths were reviewed by anaesthetistsand 3/4 of deaths reviewed by surgeons, this was unchanged from NCEPOD 1990.2

In an effort to improve local practice NCEPOD would recommend:

• Improved access to notes, especially of deceased patients.1

• More post-mortem examinations.9


47

Chap-03.qxd 2/1/02 12:04 PM Page 47

• Better communication between pathologists and clinicians.11

• Regular morbidity and mortality review meetings. Ideally these shouldbe multidisciplinary meetings to enhance the working relationships ofsurgeon, anaesthetist and physician.1

• Ensure all members of staff participate equally in audit.1

In the light of public concern over organ retention following post-mortem exam-ination there is rightly greater rigour now required for the consent to post-mortem examination. Details of the consent process are beyond the scope of thisbook. The Department of Health (DOH) has published interim guidance on consent for post-mortem examinations.14 In this guidance they also echo the recom-mendations from NCEPOD in emphasising the importance of post-mortemexamination to improving clinical care and maintaining standards.

In the 10 years that NCEPOD has reported it is clear that the rate of change isoften slow. Many of the lessons continue to be repeated and are not alwaysheeded. Both managers and clinicians need the commitment backed up withresources to implement changes in practice. In their introduction to the currentreport, Ingram and Hoile state ‘We believe that future change will depend onmoney, manpower, mentality and mentoring.’1

NCEPOD DEFINITIONS

Admission category

Elective – at a time agreed between the patient and the surgical service.

Urgent – within 48 h of referral/consultation.

Emergency – immediately following referral/consultation, when admission isunpredictable and at short notice because of clinical need.

Classification of operation

Emergency – immediate life-saving operation, resuscitation simultaneous withsurgical treatment. Operation usually within 1 h.

Urgent – operation as soon as possible after resuscitation. Operation within 24 h.

Scheduled – an early operation but not immediately life-saving. Operation usually within 3 weeks.

Elective – operation at a time to suit both patient and surgeon.

Further information

NCEPOD website: www.ncepod.org.uk


48

Chap-03.qxd 2/1/02 12:04 PM Page 48

References

1. Then and now. The 2000 report of the National Confidential Enquiry intoPerioperative Deaths. NCEPOD, London, 2000.

2. Campling EA, Devlin HB, Lunn JN. The report of the National ConfidentialEnquiry into Perioperative Deaths 1989. NCEPOD, London, 1990.

3. Quality and performance in the NHS: NHS Performance Indicators. NHSExecutive, July 2000.

4. Ingram GS. The lessons of the National Confidential Enquiry into Peri-operative Deaths. Ballieres Clin Anaesthesiol 1999; 13 (3): 257– 66.

5. Campling EA, Devlin HB, Hoile RW, Lunn JN. The report of the NationalConfidential Enquiry into Perioperative Deaths 1990. NCEPOD, London, 1992.

6. Devlin HB, Hoile RW, Lunn JN. One case per consultant surgeon or gynae-cologist. The report of the National Confidential Enquiry into Perioperative Deaths1993/1994. NCEPOD, London, 1996.

7. Campling EA, Devlin HB, Hoile RW, Ingram GS, Lunn JN. Who operateswhen? A report by the National Confidential Enquiry into Perioperative Deaths1995/1996. NCEPOD, London, 1997.

8. Campling EA, Devlin HB, Hoile RW, Lunn JN. The report of the NationalConfidential Enquiry into Perioperative Deaths 1991/1992. NCEPOD, London,1993.

9. Campling EA, Devlin HB, Hoile RW, Lunn JN. The report of the NationalConfidential Enquiry into Perioperative Deaths 1992/1993. NCEPOD, London,1995.

10. Extremes of age. The 1999 report of the National Confidential Enquiry intoPerioperative Deaths. NCEPOD, 1999.

11. Gallimore SC, Hoile RW, Ingram GS, Sherry KM. Deaths within 3 days ofsurgery. The report of the National Confidential Enquiry into Perioperative Deaths1994/1995. NCEPOD, London, 1997.

12. Gray AJG, Hoile RW, Ingram GS, Sherry KM. Specific types of surgery andprocedures. The report of the National Confidential Enquiry into PerioperativeDeaths 1996/1997. NCEPOD, London, 1998.

13. The CCST in Anaesthesia I: General Principles – A Manual for Trainees andTrainers. July 2000. The Royal College of Anaesthetists.

14. Organ Retention: Interim Guidance on Post-mortem Examination. Department ofHealth, 2000.


49

Chap-03.qxd 2/1/02 12:04 PM Page 49

Chap-03.qxd 2/1/02 12:04 PM Page 50


51

4ANALGESIA FOR THE

HIGH RISK PATIENT

In years past severe pain was accepted as an inevitable consequence of trauma andsurgery and little effort was made to provide adequate pain relief in the majority ofunfortunate patients:

• Whilst adequate pain relief is a laudable objective from the humanitarianperspective, modern understanding of the pathophysiological effects ofpain makes appropriate pain relief a primary objective in avoiding thecommon morbidities associated with surgery.

• The patient who is at ‘high risk’ either because of the trauma of theirsurgery or their poor physiological reserve therefore requires effective painrelief to avoid these potentially lethal complications.

• If this is not achieved, then these are the patients most likely to slidedown the slippery slope to critical illness.

Modern approaches to the management of acute pain rely heavily on two analgesictechniques, patient controlled analgesia (PCA) using an opioid self-administeredin small doses by the patient, and epidural analgesic techniques. At present there isno evidence supporting a reduction in morbidity using PCA. Epidural techniqueshowever have been demonstrated to confer a number of benefits1–3 and as suchwould seem to be the analgesic method of choice in the ‘high risk’ patient. Otherlocal anaesthetic techniques used occasionally by acute pain teams may also be ofbenefit. Some aspects of local anaesthetic techniques are discussed in Chapter 5.The skills of a multidisciplinary acute pain service (APS) are essential to ensureoptimal pain management is achieved in ‘high risk’ patients.

THE ROLE OF THE ACUTE PAIN SERVICE

APSs developed in response to the joint colleges’ report ‘Pain after Surgery’(Royal College of Surgeons and College of Anaesthetists 1990) which highlighted

Chap-04.qxd 2/1/02 12:05 PM Page 51

the poor record and lack of progress in postoperative pain management over the previous 50 years.4 In order to improve pain management and safely introducenew techniques onto general wards, such as PCA and epidural infusions, the reportrecommended setting up APS led by a named consultant and a specialist nursepractitioner. Services differ slightly in structure depending upon the needs of theparticular hospital but all work to the same priorities in ensuring the attainmentof certain levels of good practice by the implementation of guidelines and proto-cols supported by education programmes and by the provision of clinical supportto advise and direct patient management at ward level. In ‘high risk’ patients itmay be worthwhile, when possible, to discuss pain management with members ofthe service in advance of the event.

THE PATHOPHYSIOLOGY OF ACUTE PAIN

Acute pain results from injury or inflammation and generally has a biologicallyuseful function. This function is protective by allowing healing and repair tooccur.5 The pathophysiological effects of acute pain are summarised in figure 4.1.Many patients experience acute pain as a result of surgery.

• The effect of an anaesthetic is to lower the functional residual capacity(FRC; the volume of gas remaining in the lung at the end of normalexpiration) of the lung.

• In elderly patients or those with concurrent lung disease the FRC mayfall below the closing volume (the volume of gas in the lung below whichsmall airways begin to close) of the lung leading to areas of atelectasis.6

• This situation may be made worse by sputum retention as a result ofprolonged surgery and in such circumstances atelectasis may develop inyounger patients.


52

Risk of PE

Risk of DVT

Impaired mobilisation

Pain on movement

Hypoxia

Pneumonia

Slow return of lung function/FRC

Poor cough/ExpectorationDeep breathing

Organ failure MI Gut/Sepsis

Organ ischaemia

Increased O2 requirementsGlobal/Myocardial

Increased BP/Heart rate

Pain

Figure 4.1 – The pathophysiology of acute pain.

Chap-04.qxd 2/1/02 12:05 PM Page 52

• An adequate cough and ability to deep breathe is essential during theearly postoperative period if these effects are to be reversed.

• This cannot generally be achieved following major abdominal or thoracicsurgery without adequate analgesia and indeed the situation may worsenfurther if cough is inadequate as this will lead to further sputum reten-tion, airway closure and ultimately pneumonia.

• Hypoxaemia as a result of this process will jeopardise the function of otherorgans. Increased myocardial oxygen requirements due to the increasein heart rate and/or blood pressure seen in the patient in pain may notbe met if the patient is hypoxic.

• This may precipitate myocardial ischaemia or lead to a perioperativemyocardial infarction.

• Hepatic and renal function may be compromised and ischaemia of thegut may contribute to postoperative ileus and breakdown of the gutbacterial barriers that could lead to sepsis.

• Early mobilisation can be facilitated by good pain relief and this in turnreduces the likelihood of deep venous thrombosis and pulmonaryembolus and will reduce the likelihood of hypostatic pneumonia.

To promise perfect analgesia is inappropriate as this may be unachievable evenwith an epidural technique, thus the aims of pain management are to achieve alevel of pain with which the individual can cope without distress and which willnot hinder coughing and mobility. In addition pain relief should encourage andfacilitate rest and normal sleep patterns whilst enabling early mobilisation and theability of the patient to communicate with their carers. Ideally analgesic regimesshould take into account periods where pain intensity is increased due to thera-peutic interventions (incident pain), e.g. physiotherapy, dressing changes, etc. Thisis particularly important in patients with coronary artery disease who may developmyocardial ischaemia as a result.

RISK FACTORS IN PAIN MANAGEMENT

Site of injury

Pain that interferes with deep breathing and coughing confers the greatest risk tothe patient and therefore the anatomical site of the surgery or injury is importantwhen assessing risk. Thoracic surgery or injuries interfere most with the mechan-ics of breathing and coughing, the next most serious are upper abdominal injuriesfollowed by lower abdominal problems and then by pain in the peripheries. Whenplanning postoperative pain relief the site of surgery must be considered in conjunction with the patient’s other risk factors.

ANALGESIA FOR THE HIGH RISK PATIENT

53

Chap-04.qxd 2/1/02 12:05 PM Page 53

Co-existing medical conditions

Certain medical conditions have implications for the choice of pain management.Opioid drugs are used in many analgesic techniques and can lead to respiratorydepression. Patients with co-existing respiratory disease, morbid obesity, sleepapnoea and the elderly are the most at risk of respiratory depression from opioids.Although opioids are commonly used via the epidural route these patient groupsmay benefit greatly from the excellent analgesia that an epidural provides, particu-larly if the site of injury interferes greatly with respiratory function. The use ofnon-steroidal anti-inflammatory drugs (NSAIDs) should be avoided in patientswith a number of conditions including renal failure, peptic ulceration, asthma andcongestive cardiac failure and should be used with care in postoperative patientswho are likely to be dehydrated. Coagulation abnormalities will preclude the use of epidural techniques and, if time and circumstances permit, considerationshould be given to reversing anticoagulation to allow the use of epidural tech-niques in patients considered to be at high risk of problems associated with pooranalgesia. Care should be taken in patients with ischaemic heart disease, whilstgood analgesia protects against myocardial ischaemia the hypotension due toepidural techniques may be undesirable in the presence of a critical coronarystenosis.

THE BENEFITS OF EPIDURAL ANALGESIA IN THE HIGH RISK PATIENT

The role of good analgesia in the avoidance of morbidity is most clearly demon-strated in patients receiving epidural analgesia. Level 1 evidence (obtained from systematic review of relevant randomised controlled trials) obtained by theAustralian Working party group (NHMRC) demonstrates that postoperativeepidural analgesia can significantly reduce the incidence of pulmonary morbidity.3

A review by Buggy and Smith concluded that current evidence demonstrates thatepidural analgesia may facilitate early recovery and improved outcome by reducingthe incidence of thromboembolic, pulmonary and gastrointestinal complicationsafter major surgery.1 The potential benefits of epidural analgesia in the high riskpatient seem clear but the small risk of neurological complications and the potentialrisk of hypotension in the individual patient must be borne in mind. There is noevidence that these benefits are manifest in patients receiving parenteral opioidanalgesia.

FUNDAMENTAL PRINCIPLES OF PAIN MANAGEMENT

Pain assessment

The 1990 Report of the Working Party of the Royal College of Surgeons andCollege of Anaesthetists recommended the systematic assessment and recording of


54

Chap-04.qxd 2/1/02 12:05 PM Page 54

pain during the postoperative period.4 There is no objective measure of pain, thereport of the patient is the only yardstick. If pain is not assessed expertly and regu-larly then the analgesic regime may be inadequate. Remember that many patientstend not to complain and will tolerate quite severe pain stoically. It is importanttherefore that the patient is involved in the process of assessment. The simplesttools are single-dimensional matching pain to a visual or verbal 0–10 scale with 0 – ‘No Pain’ and 10 – ‘The Worst Pain Imaginable’.

The key points are that:

• the tool is quick and easy to use,

• the assessment is made by the patient both at rest and on movement,

• the assessment is made regularly and repeated soon after any intervention,

• the result is acted upon if the pain score is above half way up the scale.

From a therapeutic perspective patients should be comfortably able to take a deepbreath and cough and as such measurement of pain on movement, deep breathingor coughing is a more important determinant of outcome than measurement ofpain scores at rest. Individual assessments are crucial in all patients in pain to prevent the tendency towards ‘blanket’ prescribing.

Changes in the type or intensity of the pain being experienced by the patientshould be given serious consideration as this may indicate failure of the analgesictechnique, e.g. an epidural catheter falling out or becoming disconnected, or mayindicate a deterioration in the patients condition. Early identification and treat-ment of neuropathic pain should be given consideration particularly if nerveinjury is likely. Neuropathic pain is often described as ‘burning’ or ‘shooting’ andmay be elicited by minimal stimulation of the affected area. It is poorly responsiveto morphine which is commonly given in larger and larger doses if the diagnosisis missed. Therapy with carbamezepine or amitripyline is more appropriate andshould be considered.

Multi-modal analgesia

This is also referred to as ‘Balanced Analgesia’ and implies the use of two or moreanalgesic agents in combination to effect pain relief at different places along thepain pathway. Possible analgesic agents that can be used in this way are

• opioids (higher centres and spinal cord effects via opioid receptors),

• NSAIDs (peripheral nociceptors via inhibition of cyclo-oxygenase),

• paracetamol (NSAID like effects but none of the usual side effects),

• local anaesthetics (block sodium channels and hence conduction innerve fibres),


55

Chap-04.qxd 2/1/02 12:05 PM Page 55

• tramadol and clonidine (increase activity of spinal descending inhibitorypathways by decreasing re-uptake of nor-adrenaline and 5-HT in neuralsynapses).

Drug combinations should be tailored to the individual depending upon circum-stances and contraindications. The benefits of multi-modal analgesia are welldescribed, better analgesia can often be achieved with greater safety and fewer sideeffects particularly when adjuvant analgesics are used alongside opioids when ademonstrable opioid sparing effect can be seen.

INITIAL ANALGESIA IN THE HIGH RISK PATIENT

Many patients in the ‘high risk’ category will present as emergency admissionseither as a result of trauma or their disease process e.g. acute abdomen. Effectinggood analgesia quickly should be a priority in these as in all patients. Good anal-gesia in the early stages helps reduce the physiological and psychological stressesbrought about by trauma or disease and is particularly important in patients withischaemic heart disease. There is no justification for withholding analgesia tofacilitate clinical diagnosis, not even in the patients with acute abdominal pain.Oral analgesics are of little use as nausea or vomiting may be a feature and absorp-tion of the drug unpredictable. Intramuscular (IM) or better still intravenous (IV)opioids are the method of choice supplemented by parenteral, rectal or ‘melt’NSAIDs, unless contraindicated, or rectal paracetamol. In patients who are athigher risk of respiratory depression due to current or concurrent illness, the IVadministration of an opioid to achieve analgesia is favoured as it allows carefultitration of the dose against the patients response. In most patients morphine inincrements of 1–2 mg or diamorphine in 1 mg increments would be the drugs ofchoice. It is often necessary to exceed the recommended doses for these drugs asdefined in the British National Formulary, particularly if the patient has had recentexposure to other opioid drugs.

Other techniques that may be of value in this initial phase of treatment dependingupon circumstances include inhalational analgesia using Entonox which is particu-larly useful as an adjuvant if painful interventions or movement of the patient isnecessary. In some instances a simple local anaesthetic block may be of value andcan easily be performed, e.g. femoral nerve block for a femoral fracture.

Early analgesia buys time until a more considered plan can be made to control thepatient’s pain.

ANALGESIC TECHNIQUES IN THE HIGH RISK PATIENT

APSs across the country employ a number of standard techniques to effect paincontrol. These techniques include PCA, epidural infusion analgesia (EIA), patient


56

Chap-04.qxd 2/1/02 12:05 PM Page 56

controlled epidural analgesia (PCEA), algorithm controlled opioids and a numberof other local anaesthetic blocks which may be prolonged by continuous infusionvia a strategically placed catheter. To ensure patient safety these techniques need a supporting package of protocols, education and clinical supervision that only apain service can provide. If this support is not in place the general ward is not the place for a one-off epidural. For the purpose of understanding we will givehere a brief description of the important techniques listed above which are welldescribed elsewhere.

Algorithm controlled opioids

This was first described by Gould.7 The algorithm allows the oral, IM or subcuta-neous administration of morphine,usually in 10 mg doses, as regularly as every hourin response to patient need. The algorithm allows nurses greater flexibility toadminister morphine in response to pain score, if respiratory rate, level of con-sciousness and other basic physiological parameters are acceptable. In practice anumber of doses may be needed initially to achieve analgesia after which dose frequency reduces to a more ‘normal’ 3–4 hourly pattern.

Patient controlled analgesia

The principle here is that the patient self-titrates an opioid, most commonly mor-phine, in small doses, generally 1–2 mg at a time using a patient request button.Each time a dose is administered the system ‘locks out’ usually for 5 min duringwhich time the request button is ineffective. Subsequent requests, after each 5 minlockout will result in further doses. This method is excellent for maintaining anal-gesia once achieved. Pre-loading of the patient via the IV or IM routes is manda-tory to the success of the technique as using the button alone can take hours toachieve analgesia from a standing start. The patient must have the mental andphysical capabilities to understand the technique and to use the button.

Epidural infusion analgesia

Placement of a catheter into the epidural space to effect analgesia has long beenpractised in obstetric anaesthesia. The technique is now being applied in an acutepain setting. The quality of analgesia achieved is far superior to that achieved byPCA or algorithm controlled opioids. Infusion regimes vary but usually incorp-orate mixtures of bupivacaine at a concentration of 0.0625–0.15% with a lipid soluble opioid (not morphine) such as diamorphine (maximum 40mg/ml) or fentanyl (maximum 5mg/ml). Epidural opioids are more effective when used in conjunction with a local anaesthetic to produce a synergistic analgesic actionand reduce the required dose and side-effects associated with either the localanaesthetic or opioid alone. These mixtures are run at rates of up to 10 ml/hdepending upon the site of insertion. Insertion of the epidural at an appropriate


57

Chap-04.qxd 2/1/02 12:05 PM Page 57

segmental level is important as spread of drugs within the epidural space is limited.In practice hypotension due to autonomic blockade by the local anaesthetic is a far bigger problem than respiratory depression although lowering the dose of the opioid may be wise if respiratory depression is a significant patient risk factor.

Patient controlled epidural analgesia

This is a modification of EIA. The same opioid/local anaesthetic mixtures tend tobe used at similar infusion rates. The main difference is that the patient is able toself-bolus extra doses of the mixture to supplement analgesia if required. In ourpractice we allow a patient controlled bolus of 2 ml with a 20 min lockout to sup-plement background infusions of 0.125% bupivacaine with 40mg/ml of diamor-phine at up to 8 ml/h. PCEA allows greater flexibility of dose and better patientresponse to increases in pain intensity such as during physiotherapy.

OPTIMISING ANALGESIA IN THE HIGH RISK PATIENT

Choice of analgesic technique will depend upon the site of the surgery and otherpatient risk factors. The challenge is to tailor effective analgesia to each patient’srequirements applying multi-modal principles using the available techniquesalongside adjuvant analgesics. The objective is analgesia that is effective enough toavoid further deterioration in the patient’s condition as a direct result of painwhilst avoiding side effects and complications attributable to the analgesic tech-nique. In general the technique chosen should be used, if effective, until thepatient’s pain is able to be controlled on an oral analgesic combination. It is wiseto ensure that this control is possible before permanently discontinuing a tech-nique, e.g. removing an epidural catheter.

Debate still continues regarding the use of epidurals on the general postoperativeward. In our view the full benefit of epidural analgesia is only attainable if thetechnique is maintained until the point where oral pain control is achievable.Whilst an initial period in a high dependency unit (HDU) or intensive care unit(ICU) environment is desirable in the high risk patient whilst the patient is re-warmed and fluid management is optimised, it is inappropriate to discontinuea working epidural after only 24–36 h so that the patient can go back to the general ward. As few hospitals in the UK have HDU facilities that can cope withkeeping patients for 3–4 days it is necessary to set up general wards to safely manage epidurals in order to optimise the proven clinical benefits.

ANALGESIC STRATEGIES IN THE HIGH RISK PATIENT

As site of injury is a crucial factor in pain associated risk it seems sensible to dis-cuss basic analgesic strategies using this factor as a determinate of technique.


58

Chap-04.qxd 2/1/02 12:05 PM Page 58

Pain in the peripheries

Pain in the limbs has little direct effect on breathing and coughing ability, it doeshowever significantly limit movement. Analgesic objectives should be to promoteearly mobilisation to at least a sitting in chair position to minimise the chance ofhypostatic pneumonia. Standard opioid techniques such as PCA or algorithmcontrolled opioids in combination with paracetamol and/or an NSAID would bethe method of choice. A recent paper suggests that for limb injury ketorolac is aseffective as morphine, produces less side effects and greater patient satisfaction.8

The use of epidural analgesia in these patients may preclude early mobilisation dueto motor blockade and postural hypotension. Other peripheral nerve blocks maybe of value in producing analgesia in the early postoperative period. Some blocks,e.g. brachial plexus have a prolonged action often extending into the first or evensecond postoperative day and are well worth considering particularly in patientswhere avoidance of opioids is desirable.

Lower abdominal pain

This is most commonly a result of surgery. The majority of patients having surgeryof the lower abdomen or pelvis are having elective procedures, e.g. gynaecologicalsurgery and will cope very well with PCA or on the IM algorithm. Considerationshould be given to the benefits of epidural analgesia in these patients if other riskfactors exist. Morbid obesity and/or proven sleep apnoea are a clear indication forepidural analgesia as the use of opioids in these patients is fraught with risk.Supplementation of either technique with paracetamol and/or an NSAID is desir-able and if an opioid technique is planned then on-table bilateral inguinal blocksgive excellent adjunct analgesia in the initial postoperative period.9

Upper abdominal pain

Surgical incisions on the upper abdomen may well extend into the lowerabdomen as a full blown laparotomy incision. Upper abdominal incisions interferewith the mechanics of breathing far more than lower abdominal incisions. A significant proportion of patients in this category will present as emergency caseswith the possibility of concomitant sepsis, dehydration, electrolyte imbalance andother physiological deficits. There is clear evidence that epidural analgesia, EIA or PCEA confers a benefit in this patient group. It is the technique of choice inthe majority of ‘high risk’ patients but there are always situations where epidural analgesia is impossible or should be used with care. These will include:

• Patient refusal (absolute contraindication).

• Infection at the site of insertion (absolute contraindication).

• Anticoagulation (consider reversal if for elective surgery).


59

Chap-04.qxd 2/1/02 12:05 PM Page 59

• Fixed cardiac output states, e.g. aortic stenosis, hypertrophic obstructivecardiomyopathy, epidural blockade may precipitate profound cardio-vascular collapse in these patients (use with care including full haemo-dynamic monitoring and postoperative intensive care).

• Systemic sepsis, epidural blockade may contribute to cardiovascularinstability, there is also a theoretical increased risk of epidural abscessformation (need to balance risks carefully, consider siting epidural inearly postoperative period when cardiovascular instability is less profound).

Epidural analgesia can be combined with NSAID and/or paracetamol to provideimproved analgesia using multi-modal analgesia principles. Sedative drugs andparenteral opioids should be avoided if the epidural infusion contains an opioid asthis increases the risk of respiratory depression.

If an epidural is out of the question other local anaesthetic blocks might warrantconsideration. Intrapleural or paravertebral infusions are of use for unilateral incisionssuch as open cholecystectomy. A left intrapleural block has been advocated for thetreatment of pancreatitis pain.10 An infusion of local anaesthetic directly into thewound via a catheter sited during wound closure can contribute significantly topostoperative analgesia and is of value alongside standard PCA or the IM algorithm.

Thoracic pain

Pain in the thorax interferes significantly with the mechanics of breathing andcoughing. Common causes are surgery and chest trauma. A well wired sternotomywound gives surprisingly little pain but the pain following thoracotomy is verysevere. Epidural analgesia is strongly recommended for most patients having thor-acic surgery. The only grey area is in younger patients with non-malignant diseasewhere an epidural, although giving excellent analgesia, is unlikely to influence asuccessful outcome and where neurological damage would be a major disaster. In‘high risk’ patients epidural analgesia should be given a very high priority given theexclusions previously discussed. Alternative techniques which can be used along-side PCA or algorithm controlled opioids are paravertebral infusions or epiplueralinfusions in which a catheter is sited outside the pleura in the paravertebral gutterunder direct vision by the surgeon.11 Thoracic epidural analgesia has clear benefitsin the management of chest trauma and may reduce the need for ICU admission.

Pain in more than one location

This is a common problem following major trauma and although an epidural may beindicated to treat pain from chest trauma or a laparotomy it will not give effectiveanalgesia for concomitant limb fractures. One strategy is to use local anaestheticonly in the epidural infusion and allow the patient to use a standard PCA to treat


60

Chap-04.qxd 2/1/02 12:05 PM Page 60

the pain not managed by the epidural. This strategy can also be used whenepidural analgesia is inadequate due to a missed segment or when low epiduralplacement misses the top end of a surgical wound.

PROBLEMS ENCOUNTERED WITH EPIDURAL ANALGESIA

The complications of epidural catheter insertion are well described6 and includeepidural haematoma and abscess, IV injection of local anaesthetic and inadvertentdural tap. The risk of neurological complications either of a minor or major naturehas yet to be clearly defined but must be considered when balancing the risksagainst the benefits of thoracic epidurals.

Itching, nausea and vomiting are all recognised side effects of epidural opioids.Though respiratory depression is rare monitoring of sedation level is mandatory.Hypotension (systolic blood pressure less than 80 mmHg) occurs in over 25% ofour patients. Optimising fluid management is a major challenge and in the ‘highrisk’ patient this may well best be achieved initially in an intensive or highdependency care environment.

Care should be taken with the timing of prophylactic heparin injections in rela-tion to insertion and removal of epidural catheters to reduce the likelihood ofepidural haematoma.

Epidural analgesia is associated with the development of pressure sores particularlyon the heels. This can happen even in young healthy people and nursing vigilanceis essential. Debilitated patients are at higher risk of developing this complication.Strong local anaesthetic solutions administered via the epidural in theatre may bea factor in the development of these sores. A sensible precaution is to switch offepidural infusions if patients’ legs are still paralysed beyond 2 h post surgery. It canbe recommenced when the block has regressed enough to allow leg movement.This precaution also facilitates early detection of epidural haematoma.

INCIDENT PAIN IN THE HIGH RISK PATIENT

The acute pain experience is one that begins severely, immediately followinginjury, then decreases over the subsequent few days to a level that can be controlledby simple analgesics and then, in time resolving. This is not however the wholestory as overlying this general downward trend are periods when pain intensity isincreased by therapeutic interventions such as physiotherapy, trips to the X-raydepartment, dressing changes or the patients attempts to mobilise. The analgesictechnique in use may need supplementation in order to cover these episodes. Thisis most easily achieved in the following ways:

• The use of patient controlled analgesic techniques such as PCA or PCEA.Most painful interventions can be anticipated. This allows the patient to


61

Chap-04.qxd 2/1/02 12:05 PM Page 61

dose himself/herself with extra morphine or epidural top ups in prepar-ation. In this regard PCEA has distinct advantages over EIA.

• Entonox, this is a mixture of 50% oxygen in 50% nitrous oxide adminis-tered through a mask or mouthpiece via a patient activated demandvalve. In essence it is patient controlled inhalational analgesia giving shortduration analgesia acting for as long as the patient continues to breathethe entonox. As with IV PCA the system has a built in safety mech-anism to prevent overdose if used correctly. It is essential that only thepatient holds the mouthpiece so that if the patient becomes too drowsythe mask will fall away from the face. Additionally with Entonox, thereis the psychological value of distraction with the act of using the device.

Further reading

Rawal N. 10 years of acute pain services – achievements and challenges. RegAnesth Pain Med 1999; 24: 68–73.

McQuay H, Moore A, Justins D. Treating acute pain in hospital. Br Med J 1997;314: 1531–5.

Carpenter RL,Abram SE, Bromage PR et al. Consensus statement on acute painmanagement. Reg Anesth 1996; 21: 152–6.

References

1. Buggy D, Smith G. Epidural anaesthesia and analgesia: better outcome aftermajor surgery? Br Med J 1999; 319: 530–1.

2. Rodgers A,Walker N, Schug S et al. Reduction of postoperative mortality andmorbidity with epidural or spinal anaesthesia: results from overview of ran-domised trials. Br Med J 2000; 321: 1493–7.

3. National Health and Medical Research Council Report. Acute Pain Manage-ment: The Scientific Evidence, NHMRC, Canberra, 1999.

4. Pain after surgery. Report of a Working Party of the Commission on the Provision ofSurgical Services. The Royal College of Surgeons of England and the College ofAnaesthetists. London, 1990.

5. Woolf CJ. Somatic pain – pathogenesis and prevention. Br J Anaesth 1995; 75 (2):169–76.

6. Atkinson R, Rushman G, Davies N. Lee’s Synopsis of Anaesthesia, 11th edn,1993. London: Butterworth Heinemann.

7. Gould TH, Crosby DL, Harmer M et al. Policy for controlling pain after sur-gery: effect of sequential changes in management. Br Med J 1992; 305: 1187–93.


62

Chap-04.qxd 2/1/02 12:05 PM Page 62

8. Rainer HT, Jacobs P, Ng YC et al. Cost effectiveness analysis of keterolac and morphine for treating pain after limb injury: double blind randomisedcontrolled trial. Br Med J 2000; 321: 1247–51.

9. Bunting P, McConachie I. Ilioinguinal nerve blockade for analgesia afterCaesarean Section. Br J Anaesth 1988; 61: 773–5.

10. Sinatra R, Hord A, Ginsberg B, Preble L. Acute Pain: Mechanisms andManagement, 1992. London: Mosby Year Book.

11. Richardson J, Sabanathan S, Jones J et al. A prospective, randomized compari-son of preoperative and continuous balanced epidural or paravertebral bupiva-caine on post-thoracotomy pain, pulmonary function and stress responses.Br J Anaesth 1999; 83 (3): 387–92.


63

Chap-04.qxd 2/1/02 12:05 PM Page 63

Chap-04.qxd 2/1/02 12:05 PM Page 64


65

5LOCAL ANAESTHETIC TECHNIQUES

Local anaesthetic techniques are widely used in high-risk surgical patients.Theymay be used alone or in combination with general anaesthesia to provide anaes-thesia and analgesia both intraoperatively and postoperatively. A wide variety oflocal anaesthetic techniques are used varying from simple techniques such aswound infiltration through field and nerve blocks to major regional anaesthesia.Infusion systems may be used to provide prolonged anaesthesia or analgesia intothe postoperative period.

REGIONAL AND LOCAL ANAESTHETIC TECHNIQUES

There are a large number of nerve blocks described which can be used to provideanaesthesia or analgesia for procedures or painful conditions affecting many partsof the body.These are described in detail in other texts:

• These procedures can be associated with adverse events and it is impor-tant when performing a nerve block that the anaesthetist is familiarwith the anatomy both of the nerve and also of adjacent structures, thepotential adverse events specific to the procedure being performed andtakes all precautions to reduce these risks to the patient.

• Use of a nerve stimulator when appropriate will increase the potentialfor a successful block and reduce the risk of adverse effects.

Blocks have been described which can be used to provide anaesthesia or augmentanaesthesia for procedures on many areas of the body.The major advantage of localand regional techniques is that they can be used to avoid general anaesthesia forsurgery,or allow a reduction in the anaesthetic or analgesic dosage.This can reducethe risk of complications, particularly postoperative respiratory tract infections,nausea and vomiting, and pain. Cardiac complications such as hypotension may bereduced when using regional anaesthesia, but there is conflicting evidence on this.It is important, however, to remember that practice and experience are importantfactors in the success of any technique, and that a competent general anaesthetic is

Chap-05.qxd 2/2/02 12:56 PM Page 65

preferable to the serious complications of a regional technique which has gonewrong.

Adverse events from local anaesthetic techniques may be due to the technique orthe agents used:

• General risks for all techniques include the risk of local infection,haematoma and trauma to the nerve which may lead to temporary orpermanent symptoms.When a local or regional technique is being usedas the sole anaesthetic technique it is important to consider the adverseeffect of patient stress and anxiety, which can be associated withunwanted hypertension and tachycardia.

• Other risks are specific to the block which is being performed and aknowledge of these risks is important when performing any block.

Spinal and epidural anaesthesia

These techniques are very widely performed and again are associated with poten-tial adverse events. Hypotension is a significant effect, due to blockade of sympa-thetic afferents.The sympathetic afferents originate in the thoracolumbar anteriornerve roots as far as L2. Block below this level is not associated with significanthypotension, increasing block height is associated with an increasing degree ofhypotension. Spinal anaesthesia, which is of more rapid onset and produces pro-found sensory and motor block causes more hypotension than epidural anaesthe-sia which is more gradual in onset and which can be more controlled by slowadministration.The other adverse effects of epidural and spinal anaesthesia includeheadache, backache, an increase risk of pressure sores, epidural haematoma,epidural abscess and risk of cord or nerve damage. Particularly, with epiduralanaesthesia the block may be insufficient to be used as the sole anaesthetic or analgesic agent.

REGIONAL BLOCKADE AND TREATMENT THAT INTERFERESWITH COAGULATION

Patients who have clotting disorders have an increased risk of haemorrhage dur-ing local and regional blockade and therefore regional techniques are often con-traindicated in these patients. Patients with serious haematological and liverdisease, and those with severe intercurrent disease receiving thromboprophylaxisare all at increased risk of haemorrhagic complications. Many patients receivingantiplatelet treatment or anticoagulant prophylaxis now present for surgery.These treatments are particularly common in high-risk patients and the risk ofhaemorrhage must therefore be balanced against the potential benefits of the useof a regional technique in each individual patient.1


66

Chap-05.qxd 2/2/02 12:56 PM Page 66

The incidence of haematoma following regional anaesthesia is extremely low.Factors involved in reducing the risk of haematoma formation include:

• uneventful needle insertion,

• the use of smaller needle size,

• avoidance of catheter insertion where possible,

• catheter removal timed to coincide with minimal anticoagulant effect.

Most research has been into the effect of these treatments in spinal and epiduralanaesthesia, although there are haemorrhagic risks in other nerve blocks.

Many high-risk patients arrive for surgery on aspirin,which has an effect on plate-let cyclo-oxygenase (COX) and interferes with platelet agglutination. Other non-steroidal anti-inflammatory drugs (NSAIDs) also affect platelet COX, however,their effect appears to be less prolonged than aspirin:

• The effect of aspirin can persist for 7–10 days.

• The COX 2 inhibitor NSAIDs are reported to have no effect on plateletfunction.

• There have been a number of case reports published where patientshave developed haematoma when undergoing regional blockade whileon aspirin.

• Several large studies have failed to show an increased incidence ofhaematoma formation in patients on aspirin coming for regional blockadecompared to those not on aspirin and the incidence is therefore very low.

• Ideally a patient should omit aspirin prior to admission for surgery for7–10 days.

The effect of heparin on epidural anaesthesia has also been extensively studied,looking at both standard and low molecular weight heparin (LMWH).2 Therehave been reports of spinal haematoma in patients receiving intravenous heparinand risk factors include vessel puncture during needle siting, heparinisation within1 h of needle, catheter siting and concomitant aspirin therapy:

• If intravenous heparin is to be commenced following a spinal or epi-dural, then there should be a delay of at least 1 h between the insertionof the block and commencement of heparin.

• Catheter techniques can be safely used, but the catheter should beremoved at a time when heparin activity is low.

Subcutaneous heparin is also relatively safe, with few reports of spinal bleeding.Needle insertion should not be within 4 h of the last dose, and the catheter should

LOCAL ANAESTHETIC TECHNIQUES

67

Chap-05.qxd 2/2/02 12:56 PM Page 67

not be removed for at least 4 h after a dose. Patients with liver disease, receivingantiplatelet therapy, or on long-term thromboprophylaxis will require monitoringof their anticoagulant effect.

LMWH has been given in a large number of patients who undergo spinal orepidural anaesthesia.These again are associated with case reports of haematoma.LMWH alter coagulation and have a 4 h half-life of effect.At 12 h after injection,there is still antithrombotic activity with 50% of maximum anti-IXa activity.These patients have altered coagulation and it is recommended that needle place-ment should take place at least 12 h after the last dose of LMWH with a delay of2 h before the next dose:

• The risk of haematoma is increased by concomitant use of aspirin ordextran.

• Single dose spinal anaesthesia may be safer than epidural, as a smallerneedle is used and no catheter sited.

• If a catheter is sited, this should be removed at least 12 h after a dose ofLMWH, by leaving removal to 24 h after the last dose coagulation canbe normalised.

• After removal of the catheter, there should be at least a 2 h delay beforethe next dose.

AGENTS USED IN LOCAL AND REGIONAL ANAESTHESIA

Local anaesthetic agents are the most widely used agents in local and regional tech-niques, but other agents may be used to cause vasoconstriction and thus prolongtheir effect, such as adrenaline. Many analgesic agents have been used to augmentor replace the local anaesthetic, especially in spinal and epidural techniques.Theseagents may all have adverse effects on the patient which must be considered.

Local anaesthetic agents

There are a large number of local anaesthetics available.These are tertiary aminoesters or amides, which work by acting on the nerve cell and interfering with thetransfer of sodium ions across the membrane.They thus interfere with the genera-tion of action potentials.Different nerves are affected by different concentrations oflocal anaesthetics and pain, temperature and touch fibres are affected at lower con-centrations than motor fibres. However, it is not possible to produce a completesensory block in a mixed nerve fibre without producing some motor blockade.

Local anaesthetic agents affect a variety of excitatory tissues and can have effectson a variety of systems, particularly the central nervous system (CNS) and the cardiovascular system. In the CNS, inhibitory neurones are more affected than


68

Chap-05.qxd 2/2/02 12:56 PM Page 68

excitatory, and local anaesthetics can cause tremors and restlessness and seizures inoverdose.The respiratory centre can also be affected by overdose, initially causingan increase in respiration but at higher doses respiratory depression.

Cardiovascular system effects include cardiac and peripheral effects.All local anaes-thetics apart from cocaine act as vasodilators via an effect on arterioles, bupiva-caine having less effect than many other agents. Cocaine acts as a vasoconstrictor.Autonomic ganglia may also be affected by local anaesthetics via an anti-muscariniceffect. Local anaesthetics also act directly on the excitatory tissue of the heart caus-ing an increase in the refractory period, prolonged conduction time and depres-sion of myocardial excitability. This effect has been used as an advantage in thetreatment of ventricular extrasystoles, tachycardia and fibrillation with lignocaine,although lignocaine can increase the defibrillation threshold.

Local anaesthetics are metabolised in the liver. Procaine and amethocaine are alsometabolised by plasma cholinesterases.

Side effects are rare from local anaesthetic agents, but toxicity can be a major riskespecially:

• When an overdose is given.The dose to be given should be calculatedfrom the safe dosage for each patient. If agents are mixed, repeated dosesare given or for an infusion use, this must be calculated to be within safelimits.

• Injection into a vein. It is important to aspirate prior to and duringinjection of local anaesthetic and if using an infiltration technique tokeep the needle moving to avoid this.

• Injection into very vascular tissue or application over mucous mem-brane where absorption is rapid.

• Severe hepatic impairment where metabolism and excretion may bereduced.

It is important to be familiar with and prepared to manage the effects of localanaesthetic toxicity when using as part of an anaesthetic technique. Adequatemonitoring and resuscitation equipment should be available when using theseagents and toxic dosage should be avoided.

Vasoconstrictor agents

Local anaesthetic agents may be used in combination with a vasopressor. These are used to counteract vasodilatation, which will slow systemic absorption of theagent and reduce bleeding in the surgical field. By slowing systemic absorption of the agent, it is possible to increase the safe dose of many but not all localanaesthetic agents and also to prolong their effect.


69

Chap-05.qxd 2/2/02 12:56 PM Page 69

Adrenaline is most widely used as a vasopressor agent in combination with localanaesthetics. Despite acting as a vasoconstrictor, it is systemically absorbed and canproduce a significant effect causing tachycardia, hypertension and anxiety. In dental anaesthesia, this has been shown to have a significant effect in elderlypatients with hypertension who demonstrate an increased risk of tachycardia andhypertension.3 However, it has been used safely for many years in these patientsand its effect on morbidity and mortality seems to be minimal. Felypressin is analternative vasoconstrictor used in dental practice which avoids this effect.

Many other drugs have been used in combination, either alone or in combinationwith local anaesthetic to provide enhanced analgesia and anaesthesia. In combina-tion with low dose local anaesthetic, they can provide intra- and postoperativeanalgesia with less risk of adverse effects from use of one or other alone agent.These drugs are used commonly in as adjuvants in spinal and epidural anaesthesia,but are also effective in plexus and nerve blocks.

Opioid analgesics

These are used especially in spinal and epidural techniques to enhance the anal-gesic effect and also to allow a reduction in the concentration and dosage of localanaesthetic.They can be given in much smaller doses than their systemic dose andare less likely to produce respiratory depression and other side effects. Opioidsgiven spinally work locally on the opioid receptors in the dorsal horn, but alsospread more centrally within the CNS. For this reason, respiratory depression canoccur, especially following intrathecal opiate administration and this may be a latecomplication.The rate of onset of spinal opioids is related to their lipid solubility.

Epidural opioids work both at local spinal receptors but also at supraspinal levels.Epidural opioids are absorbed into the epidural venous plexus and this is probablya major contribution to their supraspinal effect.When given epidurally, the effectis related to the lipid solubility of the opioid used:

• Lipophillic agents such as fentanyl are rapidly absorbed by surroundingfatty tissues as well as crossing the dura, thus having a rapid onset butrequiring higher doses than less lipid soluble agents such as morphine.

Respiratory depression is a major complication of spinal and occasionally epiduralopioids, and may occur several hours following administration. It appears to be a supraspinal effect and can be reversed by naloxone.Pruritis and urinary retentionare other side effects of intrathecal and epidural use of opiates.

Other agents

These are used to augment regional anaesthesia and include clonidine, an alphaagonist.Clonidine has been used spinally and epidurally and also as part of brachialplexus and lower limb nerve blocks. It is used in combination with local anaesthetic


70

Chap-05.qxd 2/2/02 12:56 PM Page 70

to potentiate the effect and duration of the block:

• Clonidine is associated with hypotension and drowsiness at low dose,which could preclude its use in many high-risk patients.

• It causes less urinary retention than opiates when used spinally andepidurally.4

• When used in nerve blocks, clonidine prolongs and augments the localanaesthesia without significant hypotension.

BENEFICIAL EFFECTS OF REGIONAL BLOCKADE ON THESTRESS RESPONSE TO SURGERY

There is an endocrine response to surgery which causes an increase in the pro-duction of pituitary hormones and stimulation of the sympathetic nervous system.This process produces a number of metabolic effects which are generally catabolic.

The stress response is considered a contributor to patient morbidity and mortalityfollowing major surgical procedures, and ways of reducing the response havetherefore been widely studied. In particular, the use of regional anaesthetic tech-niques and especially epidural anaesthesia have been studied in relation to theirmodification of the stress response.5 This has, so far, not been extrapolated to anoverall increase in survival in any study. See also Chapter 12.

BENEFITS OF SPINAL AND EPIDURAL ANAESTHESIA

These have been extensively studied and shown to have a number of potentialbenefits for the patient:

• Thromboembolic effect – Regional anaesthesia has been shown toreduce the incidence of thromboembolic complications considerably.This is especially the case following surgery to the lower limbs andpelvis.The effect is partly due to lower limb vasodilatation, inhibition of the stress response and therefore reduced platelet aggregability,alteration of tissue aggregation and inhibition factor production.6

• Cardiovascular effect – There is a reduction in hypertensive andtachycardic response to surgery with regional anaesthesia.There may bea reduction in risk of myocardial infarction or cerebrovascular accident(CVA) for patients undergoing regional anaesthesia.7

• Respiratory – There is an improvement in respiratory parameters inpatients receiving regional anaesthesia compared to general anaesthesia.There is also a reduction in postoperative respiratory infections.Patientswith severe chronic lung disease have been shown to suffer less post-operative respiratory complications following regional anaesthesia.8


71

Chap-05.qxd 2/2/02 12:56 PM Page 71

• Blood loss – The transfusion requirements of patients undergoing aregional anaesthetic technique are reduced both intra- and postopera-tively when a regional technique is used.

• Gastrointestinal function – This is improved by regional anaesthesiawith local anaesthetic and opioid following abdominal surgery.There is a lower incidence of ileus compared to patients receiving systemic opioids or epidural opioid alone.This allows a return to early enteralnutrition and the postoperative benefits of improved nutrition for thepatient.

• Infection – There is a reduction in overall infection rates.This is par-ticularly the case for respiratory tract infections.

• Postoperative recovery – There is an improvement in patient analge-sia with regional techniques.

• Mortality rate – Some studies have shown a reduction in mortalityrates for patients undergoing regional anaesthesia, with or withoutgeneral anaesthesia, however, this is not universally reported and somestudies have shown a possible increase in mortality rates.

The advantages listed above have been hoped to be associated with a demonstra-tion of shortened hospital stay and postoperative complications, but these have notbeen proven conclusively.

LOCAL AND REGIONAL ANAESTHESIA IN PATIENTS WITHSERIOUS INTERCURRENT DISEASE

The theoretical advantages of avoiding general anaesthesia and providing a theo-retical reduction in cardiovascular and respiratory risks has made local and regionalanaesthesia very popular in patients with serious coexisting disease. Groups whohave been extensively investigated include the following.

Patients with cardiac disease

These patients are at risk of increased perioperative morbidity and mortality, andthis is especially the case in those with recent myocardial infarct or congestive car-diac failure. Patients at risk of cardiovascular complications include patients withknown ischaemic heart disease, hypertensive disease and the elderly.

Local and regional techniques are often used in this group to avoid the cardiovas-cular effects of general anaesthesia and they have theoretical advantages, e.g. thereduction in the hypertensive and tachycardic response to surgery.

Local anaesthetic techniques for dental and eye surgery in these patients have been used very safely.The use of adrenaline as an adjunct to dental anaesthesia has


72

Chap-05.qxd 2/2/02 12:56 PM Page 72

been performed safely for many years without serious clinical problems in thisgroup of patients, although there is an increased risk of arrhythmias especially in those with congestive cardiac failure or taking digoxin.9There is a risk of anxiety-related problems in these patients which can counteract some of the cardiac benefitsof avoidance of general anaesthesia.

Local and regional anaesthesia has also increasingly replaced general anaesthesia for carotid endarterectomy. There may be an advantage related to incidence ofmyocardial ischaemia for local anaesthesia compared to general anaesthesia but thisis, so far, not proven.10

Spinal and epidural anaesthetic techniques are also used widely in these patients.There is evidence that epidural anaesthesia is safer than spinal anaesthesia:

• Spinal anaesthesia is associated with more hypotension than epiduralanaesthesia and this is of rapid onset.

• In patients with cardiac risk, use of spinal anaesthesia has also been asso-ciated with an increased incidence of postoperative cardiac complica-tion and mortality in comparison to general and epidural anaesthesia.

However, the overall message is that a recent large review of regional anaesthesiasupports the view that spinal and epidural techniques are associated with reducedmortality compared to general anaesthesia used alone.11

Patients with severe pulmonary disease

This group of patients has an increased risk of postoperative respiratory complica-tions and local anaesthetic techniques are often used as the treatment of choice inthese patients:

• General anaesthesia has been reported as a risk factor for the develop-ment of a postoperative respiratory complication when compared withlocal or regional techniques.

• Patients with respiratory disease presenting for orthopaedic surgeryhave been shown to have reduced morbidity and mortality followingregional techniques.

• There is controversy still over whether use of postoperative regionalanalgesia can reduce pulmonary mortality following major surgery,12

although it appears likely to have a beneficial effect.

Patients with chronic renal failure

Local and regional techniques are both used to provide anaesthesia for formationof arterio–venous fistulae in this group of patients.Brachial plexus blocks have beenshown to be very safe in these patients and are used widely for fistula formation.


73

Chap-05.qxd 2/2/02 12:56 PM Page 73

Using a regional anaesthetic technique should also provide these patients withbeneficial effects due to avoidance of the effects of general anaesthesia on renalblood flow and the electrolyte and water retention produced by the stressresponse. However, it is important to maintain a normal blood pressure and cardiac output to retain this benefit.13

Elderly patients

These patients frequently have a number of medical conditions that increase theirperioperative risk and therefore frequently receive regional or local anaesthesia.Studies have shown beneficial effects from regional anaesthesia, particularly inorthopaedic surgery.

Orthopaedic surgery

Regional anaesthetic techniques are used widely for patients undergoing ortho-paedic and trauma surgery.The benefits of a reduction in thromboembolic com-plications, haemorrhage, and postoperative infections have led to their use in thesepatients who are often at high risk due to their age and intercurrent medical prob-lems.There is an early reduction in morbidity and mortality, although this does not always extrapolate to long-term improved survival and some studies have been unable to show any relative benefit for regional anaesthetic techniques overgeneral.14

Further reading

Atanassoff PG. Effects of regional anesthesia on perioperative outcome. J ClinAnesth 1996; 8: 446–55.

Hall GM,Ali W.The stress response and its modification by regional anaesthesia.Anaesthesia 1998; 53 (suppl. 2): 10–12.

References

1. Knowles PR. Central nerve block and drugs affecting haemostasis – are theycompatible? Curr Anaesth Crit Care 1996; 7: 281–8.

2. Horlocker TT, Heit JA. Low molecular weight heparin: biochemistry, phar-macology, perioperative prophylaxis regimens and guidelines for regionalanesthetic management. Anaesth Analg 1997; 85: 874–85.

3. Zhongxiang-Lim, Wengi-Geng. Tooth extraction in patients with heart dis-ease. Br Dental J 1991; 170: 451–3.

4. Gentili M, Bonnet F. Spinal clonidine produces less urinary retention thanspinal morphine. Br J Anaesth 1996; 76: 872–3.


74

Chap-05.qxd 2/2/02 12:56 PM Page 74

5. Hall GM, Ali W. The stress response and its modification by regional anaes-thesia. Anaesthesia 1998; 53 (suppl. 2): 10–12.

6. Neimi TT, Pitkanen M, Syrjala M, Rosenberg PH. Comparison of hypoten-sive epidural anaesthesia and spinal anaesthesia on blood loss and coagulationduring and after total hip arthroplasty. Acta Anaesth Scand 2000; 44: 457–64.

7. Rodgers A, Walker N, Schug S et al. Reduction of postoperative mortality and morbidity with epidural or spinal anaesthesia: results from overview ofrandomised trials. Br Med J 2000; 321: 1493–7.

8. Pedersen T. Complications and death following anaesthesia. Danish Med Bull1994; 41 (3): 319–31.

9. Blinder D,Shemesh J,Taicher S.Electrocardiographic changes in cardiac patientsundergoing dental extraction under local anaesthesia. J Oral Maxillofacial Surg1996; 54 (2): 162–5.

10. Ombrellaro M, Freeman MB, Stevens SL, Goldman MH. Effect of anesthetictechnique on cardiac morbidity following carotid artery surgery. Am J Surg1996; 171: 387–90.

11. Racle JP, Poy JY, Haberer JP, Benkhadra A. A comparison of cardiovascularresponses of normotensive and hypertensive elderly patients following bupiva-caine spinal anesthesia. Reg Anesth 1989; 14 (2): 66–71.

12. Ballantyne JC et al. The comparative effects of postoperative analgesic ther-apies on pulmonary outcome: cumulative meta-analyses of randomised,controlled trials. Anesth Analg 1998; 86 (3): 598–612.

13. Burchardi H, Kaczmarczyk G. The effect of anaesthesia on renal function.Eur J Anaesthesiol 1994; 11: 163–8.

14. Gilbert TB et al. Spinal anaesthesia versus general anaesthesia for hip frac-ture repair: a longitudinal observation of 741 elderly patients during 2 year follow-up. Am J Orthopaed 2000; 29 (1): 25–35.


75

Chap-05.qxd 2/2/02 12:56 PM Page 75

Chap-05.qxd 2/2/02 12:56 PM Page 76


77

6THE CRITICALLY ILL PATIENT IN THE

OPERATING THEATRE

Anaesthetic management of the critically ill patient who requires operative inter-vention remains a significant challenge.

These patients should always be anaesthetised by senior anaesthetists.

Such patients present to the anaesthetist from three main areas within the hospital:

1. The accident and emergency departmentVictims of major trauma requiring immediate operative intervention fallinto two categories:

• Major haemorrhage of any source that cannot be controlled by simpleresuscitative measures such as pressure dressing and splinting may trans-fer to the operating theatre while active fluid resuscitation is ongoing.

• Patients with traumatic intracranial haemorrhage resulting in increasedintracranial pressure (ICP) will need urgent decompression if they areto avoid medullary ‘coning’. Again such casualties may require opera-tive intervention prior to instituting full resuscitative measures.

Patients presenting with acute general surgical pathology of a non-traumatic nature may occasionally proceed from the accident andemergency (A&E) department straight to theatre, however, it is far morelikely that there will be adequate time for some degree of resuscitationand detailed investigation on the general ward or high dependency unit(HDU) prior to surgery.

2. The hospital ward or high dependency unitPatients who have already been admitted to the ward environment maydeteriorate during the course of their management. This may necessi-tate a more precipitous trip to the operating theatre than had originallybeen anticipated. It is, however, likely that a degree of resuscitative inter-vention will already have occurred.

Chap-06.qxd 2/2/02 12:59 PM Page 77

3. The intensive care unitThis group of patients have the advantage to the anaesthetist that, pro-vided they have spent a number of hours on the unit, they are mostlikely to have all resuscitative measures in place. Mechanical ventilationhas usually been instituted, together with invasive lines for both moni-toring and the administration of drugs and fluid.

Clearly the corollary of this situation is that this group of patients may be pro-foundly ‘sick’ receiving multi-system support on the intensive care unit (ICU);support that should ideally continue during any trip to the operating theatre.

These then are the patients who may fall into the category of critical illness.Regardless of the source of both patient and surgical pathology the issues andprinciples of anaesthesia surrounding any operative procedure on them remain thesame. The individual patient and his or her pathology merely alter the emphasis.The remainder of this chapter will consider these principles in some detail.

PATIENT TRANSFER TO AND FROM THE OPERATINGTHEATRE

The safe transfer of any patient around a hospital requires organisation and plan-ning. Even the most urgent of transfers to the operating theatre must not beundertaken, until all steps to ensure that the patient will not be harmed by thetransfer have been addressed. One needs to guard against complacency becauseone is ‘only going down the corridor’.

The principles of safe patient transfer are the same regardless of the distanceinvolved. There are a number of texts devoted to this topic. The Association ofAnaesthetists of Great Britain and Ireland and the Intensive Care Society havepublished guidelines for safe patient transfer (see Further reading). These include:

• Patient’s airway must be adequately secured.

• Ventilation must be adequate, either spontaneous or mechanical. It hasbeen shown that manual ventilation with a bag is unpredictable andunreliable compared with a portable mechanical ventilator. Ventilate withthe same modes as in ICU. Modern portable ventilators can supply posi-tive end expiratory pressure (PEEP) vary the inspiratory : expiratory (I : E)ratio and provide other modes of respiratory support.

• Lifting of patients on and off trolleys is a cause of inadvertent extub-ation. It is probably safest to temporarily disconnect the ventilator for afew seconds during movement.

• Blood pressure (BP) must be maintained with a combination of fluidsand inotropic agents. Stabilise patient before transfer, if possible.

• Patient monitoring must be appropriate to ensure safe transfer.


78

Chap-06.qxd 2/2/02 12:59 PM Page 78

• Consideration should be given to pharmacological sedation and musclerelaxation as indicated by the clinical condition.

• Communication between transferring and receiving staff should ensuresafe receipt of the patient.

• One must avoid last minute panic and rush. Planning should be such asto minimise delays and waiting in theatre reception areas. Check theavailability of equipment in the X-ray department before the transfercommences. Check that adequate porter services are available.

• Appropriate equipment required during the transfer includes a portableventilator, full oxygen cylinder, equipment for reintubation, drugs – forexample, sedation, paralysis, cardiac resuscitation, self-inflating bag orequivalent in the event of ventilator/oxygen supply failure and battery-powered syringe pumps if required. There is no excuse for battery-powered equipment becoming exhausted, oxygen cylinders emptyingor drug syringes running out.

Pitfalls and problems

• Inadequate resuscitation. Beware of occult injuries in multiple traumapatients.

• Staff and equipment problems. Inexperienced medical or nursing staffshould not be used for transferring critically ill patients.

• Appropriate technical support should be available and take responsibilityfor the necessary equipment.

As important as ensuring the safety of the patient to be transferred is the import-ance of not delaying the transfer to the operating theatre by undertaking proced-ures that can be performed later during the operation. For example, if a patient isexsanguinating and needs a laparotomy for abdominal trauma, there is little to begained by spending time in the A&E department inserting an arterial line. Thisprocedure can be performed during the laparotomy when the surgeon has begunto effect haemostasis. There is no merit in delivering a corpse with an arterial lineto the operating table.

PATIENT POSITIONING

When positioning the critically ill patient there are a number of points that meritemphasis:

• The critically ill patient rarely travels alone! The number of lines, tubesand bags increases with the severity of the patient’s condition. Everypiece of equipment inserted into the patient is there for a reason

THE CRITICALLY ILL PATIENT IN THE OPERATING THEATRE

79

Chap-06.qxd 2/2/02 12:59 PM Page 79

(or time should not have been wasted inserting it) and it, therefore,mustbe accessible during an operative procedure.

• Patients who have come to theatre as a result of trauma may well nothave had a full primary and secondary survey (as the operative proced-ure may constitute ‘C’ of the primary survey). In such cases it is vitalthat the presence of as yet undiagnosed fractures to any part of the spineis taken into account when moving and positioning the patient. In particular the cervical spine should stay fixed with head blocks andstrapping and the patient should not be moved without formal logrolling technique being used.

• Patients who have been critically ill on the intensive care and have asignificant sequestration of fluid into the extra-vascular compartmentswill have oedematous skin that is weakened and is prone to tearing,bruising and vulnerable to pressure injury. Every effort should be madeto minimise any damage done to the skin in such cases by providingadequate support and padding to the patient’s exposed extremities.

PERIOPERATIVE HYPOTHERMIA

Maintenance of body temperature is important. Although there is some limitedevidence that heat generation may occur following certain types of acute injury it is far more common for the traumatised patient to present to the operating theatre cold and peripherally ‘shut down’. The reasons for this are as follows:

• Following acute blood loss the cardiovascular response is profound peri-pheral vasoconstriction resulting in maintained perfusion of vital organs,brain, heart, lungs and kidneys at the expense of other vascular beds.

• During acute traumatic injury central mechanisms of thermoregulationare disrupted. Thus shivering is diminished or absent. Whether this issecondary to reduced oxygen delivery or a response to altered hormonalactivity in the thermoregulatory centre in the brain stem is unclear.

• In order to fully assess the extent of injury in the traumatised patient it is necessary to remove clothing and leave the patient exposed duringrepeated examination. This is compounded by the infusion of unwarmedintravenous (IV) fluid and blood worsening the relative hypothermia.

In addition to the above problems in trauma patients, all patients undergoingmajor surgery are at risk of becoming hypothermic (core temperature � 36�C).Reasons include:

• reduced metabolic rate associated with anaesthesia,

• vasodilation under anaesthesia,


80

Chap-06.qxd 2/2/02 12:59 PM Page 80

• abolished subclinical shivering,

• exposure,

• cold fluids used for skin preparation – which are usually allowed toevaporate,

• inadequately warmed IV fluids.

Adverse effects of perioperative hypothermia

Postoperative hypothermia has become recognised in recent years as a significant,and common problem:

• Delayed awakening due to decreased clearance of anaesthetic agents.

• Most organ function is depressed by hypothermia.

• Haemodynamic instability during rewarming – increased fluids oftenneeded as the patient vasodilates during rewarming. The hypotensionthus produced can be confused with continued bleeding.

• Oxygen consumption is increased by about 140% by shivering duringrewarming. If oxygen delivery to the tissues is not able to match thisincrease, the oxygen debt is prolonged.

• Wound infection rates may be increased by reductions in skin bloodflow.

• Cell-mediated immune function may be reduced.

• Hypothermia causes coagulopathy and a decrease in platelet count.Intra- and postoperative blood loss is increased with hypothermia, forexample, the typical decrease in core temperature during hip replace-ment increases blood loss by about 500 ml.1 Normalisation of clottingproblems will require normalisation of temperature as well as givingclotting factors.

• Adrenergic responses are increased postoperatively in hypothermicpatients – responsible for increased cardiac morbidity. There is a 55%less relative risk of adverse cardiac events when normothermia is main-tained.2 Unintentional hypothermia is associated with increased inci-dence of myocardial ischaemia in the postoperative period.

Note:

1. The degree of hypothermia in many of the studies cited was not that severe � 35°C. Thus, development of hypothermia after prolonged sur-gery is highly significant and warrants serious attention to its preventionand management.


81

Chap-06.qxd 2/2/02 12:59 PM Page 81

2. Laboratories perform coagulation studies at 37°C – regardless of thetemperature of the patient at the time the sample was taken. Thus, thesestudies may underestimate the degree of impairment of coagulopathy inthe hypothermic patient – what is after all a dynamic problem in vivorather than in vitro.

Prevention of hypothermia

All practical measures should be undertaken to minimise heat loss and maintainthe patient’s body temperature:

• Circle system ventilation with carbon dioxide absorber and heat andmoisture exchanger in the patient circuit.

• Fluid warmer for all IV fluids.

• Warmed patient mattress.

• Insulation of all areas of the patient that do not need to be exposed foreither surgical or anaesthetic access. This may be achieved by wrappingor the application of an air warming system.

VENTILATION AND AIRWAY MANAGEMENT

Most critically ill patients presenting to the operating theatre will already havesome form of definitive airway control in place. Under most circumstances itwould be prudent to leave this airway alone for fear of losing control in a patientwho may have acquired abnormalities with their airway due to tissue swelling ortrauma. If the airway is not secure, one should assume a full stomach and takeappropriate precautions – assume a cervical spine injury in all trauma patients.Under certain circumstances it is appropriate to use the trip to the operating the-atre as an opportunity to alter airway management. For example, patients whorequire ventilation on the ICU for an extended period benefit from the insertionof a tracheostomy. Although it is often possible to do this via the percutaneousroute in the ICU, on occasions where technical difficulties preclude this it may bepossible to combine an operative event in theatre with insertion of a tracheostomy,thus limiting patient transfers.

Ventilation of the critically ill patient should always be controlled using appropri-ate drugs for anaesthesia and muscle relaxation. There is no place for spontaneousventilation in this circumstance. Controlled mechanical ventilation not onlyallows surgical access, it may allow optimisation of gas exchange by manipulationof minute volume and oxygenation.

Where at all possible the ventilatory strategy undertaken should attempt to avoidvolutrauma and barotrauma both of which may serve to worsen any degree of


82

Chap-06.qxd 2/2/02 12:59 PM Page 82

acute respiratory distress syndrome (ARDS) from which the patient may be suf-fering. Ideally the mode of ventilation in the operating theatre and, indeed, intransit to and from the ICU or A&E should be of the same standard as can bedelivered in the ICU. Pressure control ventilation with the ability to alter (reverse)the I : E ratio and to apply PEEP is ideal. Lack of ongoing ventilation with PEEPand the other lung recruitment manoeuvres taken in the ICU will result in loss of recruitment of alveoli and hypoxia. This should not be a problem for most ‘normal’ patients ventilated in theatre as part of anaesthesia but critically illpatients under anaesthesia are different:

• Most critically ill patients presenting for anaesthesia have significantacute respiratory disease.

• Preoperative presence of increased pulmonary vascular resistance iscommon in critically ill patients.

• The usual presence of diseased lungs with lesser compliance results ingreater increases in peak and mean airway pressures compared to ‘normal’ patients under anaesthesia.

Occasionally to ensure minimal deterioration in respiratory physiology it may be necessary to move a static ICU ventilator to the operating theatre and ventilatethe patient on it throughout the procedure. Under this circumstance it would benecessary to adopt a total IV anaesthetic technique.

There has for some time now been a need for transport ventilators capable ofdelivering appropriate modes of gas delivery for critically ill patients. Recently anumber of genuinely portable machines with these facilities have become available.

CHOICE OF ANAESTHETIC AGENTS

Every available technique and drug combination has been used to anaesthetise thecritically ill patient. To some extent the reader must distil his or her own techniquefrom the many approaches that they will see during their training. The way a drugis used, for example, dose, speed of injection, etc. may be more important in manypatients than the absolute choice of drug. That being said, there are, however,a number of pharmacological properties and principles that should aid in thisdecision-making.

Induction agents

1. Thiopentone (sulphur analogue of pentobarbitone): Remains the mostrapidly effective drug for the induction of general anaesthesia. Rapidsequence induction for the purposes of securing the airway is best performed with Thiopentone. Thiopentone is the only induction agentthat reduces the brain’s metabolic requirement for oxygen and is hence


83

Chap-06.qxd 2/2/02 12:59 PM Page 83

neuroprotective. It produces depression of myocardial contractilitytogether with vasodilation resulting in a fall of BP. However, it is note-worthy that in the hypovolaemic or shocked patient the sleep dose ofThiopentone is greatly reduced as compared to the healthy patient.

2. Etomidate (carboxylated imidazole): The main advantage is said to becardiovascular stability when compared to Thiopentone. It inhibits 17�,11� hydroxylase enzymes in adrenal steroid synthesis. Therefore, it is contraindicated as an infusion due to the above adrenal suppressionresulting in increased mortality in patients sedated with this agent.

3. Propofol (di-isopropyl phenol): Rapid onset (although a little slowerthan Thiopentone) and obtunds pharyngeal and glottal reflexes to agreater extent than Thiopentone. Widely used for total IV anaesthesiaand ICU sedation. The hypotension produced is chiefly secondary tovasodilation rather than myocardial depression.

4. Ketamine (phencyclidine derivative): Sympathomimetic effects main-tain BP but the increases in heart rate (HR) and stroke volume increasemyocardial work. It is profoundly analgesic and is an effective analgesicagent at subanaesthetic doses. Despite sympathomimetic effects, it maycause cardiac depression, myocardial ischaemia and collapse in shockedpatients in whom catecholamine stores may be exhausted.

Opioids

The key issue is the cautious use of short acting agents. The shortest available isRemifentanil.

Advantages:

• very short acting (metabolised by plasma cholinesterase),

• no accumulation in hepatorenal failure,

• very potent,

• suitable as an adjunct to sedation,

• may reduce the need for muscle relaxants.

Disadvantages:

• must be reconstituted from powder,

• profound respiratory depression,

• must be infused,

• no postoperative analgesia,


84

Chap-06.qxd 2/2/02 12:59 PM Page 84

• causes decrease in BP (decrease in SVR and myocardial contractility) –worsened in hypovolaemia.

The lack of postoperative effect of Remifentanil (said to be like ‘liquid nitrousoxide’) and its titratability cause many anaesthetists to prefer its use for all high riskpatients. Of course, additional measures must be taken for postoperative analgesia,for example, epidural anaesthesia.

Other opioids

• Fentanyl: minimal effect on the cardiovascular system in the stable, calmpatient undergoing cardiac surgery. In shocked patients exhibiting highsympathetic tone, abolition of this with Fentanyl results in a fall in BP.

• Morphine should be used with great care in the critically ill patient asit has pharmacologically active metabolites and is reliant on the liverand kidneys for its elimination.

Muscle relaxants

There are four main factors governing the choice of muscle relaxant for anaesthesia:

• Onset – All critically ill patients are assumed to have a full stomach.Despite recent introduction of faster onset non-depolarising drugs suchas Rocuronium, Suxamethonium remains the ‘gold standard’ for rap-idly securing the airway. If cardiac instability is a major concern,Rocuronium may be a better choice.

• Cardiovascular effects – Rocuronium has least cardiac effects of the relax-ants followed by Vecuronium. However, the vagolytic and sympatho-mimetic effects of Pancuronium may make it an appropriate choice inshocked patients.

• Termination of effect and excretion – Agents not dependent on the kidneyor liver for termination of effect sound attractive in the shocked patientbut from a practical point of view few critically ill patients are ‘reversed’at the end of the operation and, thus, length of action of the musclerelaxants is not a big problem.

• Duration – In a similar manner, short or long duration is not usually an issue.

Inhalational agents

1. Enflurane – Greatest degree of myocardial depression for equivalentMAC.

2. Sevoflurane – Less increase in cerebral blood volume. Short acting.

3. Halothane – Rarely used nowadays. Long acting with more activemetabolites retained in body than other volatile agents with potential


85

Chap-06.qxd 2/2/02 12:59 PM Page 85

for liver toxicity. Sensitises the heart to endogenous and exogenous catecholamines with the potential for arrhythmias.

4. Isoflurane – Impressive safety profile in large numbers of patients.Hypotension chiefly by vasodilation rather than myocardial depression.Early concerns re-coronary steal are unfounded in conventional usage.

5. Desflurane – Specialised delivery systems required. Short acting. Someconcerns re-coronary steal.

6. Nitrous oxide – In normal patients mild indirect sympathetic stimula-tion reduces any myocardial depressant actions of N2O. Followinghaemorrhage this protecting effect is lost and N2O has the same depres-sant effects on the heart as Halothane. With the additional concerns re-lesser inspired oxygen concentrations and the potential for expansionof air spaces, for example, pneumothoraces, it is difficult to see a majorrole for N2O in critically ill patients.

Adverse effects of anaesthesia in shocked patients

In addition to the usual adverse effects of anaesthesia the critically ill and high risksurgical patient may be at additional risk from exposure to anaesthesia:

• Anaesthesia modifies the normal compensatory response to hypoxia inanimals.3 The normal compensatory increase in cerebral and coronaryblood flow does not occur under volatile anaesthesia. Thus, in the crit-ically ill patient who becomes hypoxic, anaesthesia potentially furthercompromises oxygenation of the vital organs.

• Again in animals,Enflurane attenuates the sympathetic responses to haem-orrhage resulting in worse haemodynamics than the non-anaesthetisedstate.4

Choice of anaesthetic agent in the shocked patient

Controlled studies on shocked patients undergoing anaesthesia are problematicdue to

• differences in severity of injury or shock,

• differences in fluids administered,

• adequacy of resuscitation prior to surgery,

• haemodynamic state and degree of cardiovascular support,

• previous health most notably cardiac reserve.

Thus one must take guidance from basic anaesthetic and pharmacological prin-ciples including the principles summarised above. In addition, studies on animal


86

Chap-06.qxd 2/2/02 12:59 PM Page 86

models are available, case studies and series may be of interest and there are reportson the use of anaesthesia in military situations.

A major dilemma is how to provide any anaesthesia for the profoundly shockedpatient. Thiopentone is said (almost certainly incorrectly) to have killed moreAmericans at Pearl Harbor than the Japanese! Many patients suffering from anexsanguinating injury may be so ‘shocked’ as to be thought to not require or beable to survive any anaesthetic administration – what used to be known as theoxygen and Pancuronium anaesthetic! While the intent to save life in this situationis laudable the absence of recordable BP does not guarantee lack of awareness. It isstrongly recommended that, at the very least, small amounts of Midazolam aregiven to the patient as this will reduce the incidence of recall.5 As the patient’scondition improves, for example, as haemorrhage is controlled judicious amountsof opiods and other anaesthetic agents may be introduced.

Ketamine may be a useful option in the above circumstances but the profoundlyshocked patient whose endogenous catecholamine stores have been exhaustedmay still suffer profound falls in BP on induction. Indeed, Ketamine has negativeinotropic effects on human heart muscle in vitro and reduces the heart’s ability torespond to � stimulation.

There are some animal studies to guide choice of anaesthesia in the shocked patient:

• Ketamine was associated with significantly increased survival comparedwith other agents in a model of haemorrhagic shock.6 In that study theanimals anaesthetised with Ketamine had better preservation of cellstructure in the splanchnic organs.

• Ketamine was associated with increased cardiac output than Thiopentonein another animal haemorrhagic shock model.7 Vital organ blood flow was also improved in the Ketamine group. The percentage of blood volume loss required to cause significant hypotension was signifi-cantly less in the Thiopentone group.

• In one of the few studies in high risk patients, low dose Ketamine preserved cardiac function and myocardial oxygen balance comparedwith Thiopentone.8

• In critically ill patients the use of Ketamine is more unpredictable.Most patients increase cardiac output and BP but a small percentagedemonstrate falls in cardiac parameters.9

Choice of anaesthetic agent in the septic patient

There are no controlled studies in septic patients from the ICU undergoing surgeryin the operating room. What guidance is available comes from case reports andanimal studies. Total IV anaesthesia with Propofol (and more recently Remifentanil)


87

Chap-06.qxd 2/2/02 12:59 PM Page 87

has been poorly studied in these patients although many anaesthetists have experi-ence of continuing Propofol sedation from ICU into theatre:

• The sympathetic stimulation associated with the use of Ketamine mayresult in improved haemodynamics, diuresis and reduction in the degreeof cardiovascular support.10

• In animals with septic shock, volatile agents are associated with increasesin serum lactate while Ketamine was associated with reductions in lac-tate. Ketamine preserved SVR and BP best.11 To summarise a complexpaper,Ketamine best preserved cardiac function and tissue oxygenation.

• In an ARDS model there is an increase in the inflammatory responseand immediate mortality with volatile anaesthetic agents compared withKetamine.12

Thus, animal studies and case reports strongly support the use of Ketamine inshocked and septic patients undergoing anaesthesia but caution is still advised.Unfortunately there is a shortage of convincing comparative patient studies andthe above animal studies are not necessarily directly transferable into clinical prac-tice. Most anaesthetists use the techniques they are most familiar with, either totalIV anaesthesia or inhalational anaesthesia, for the critically ill patient in the oper-ating theatre. Few have much experience of Ketamine in the UK. Therefore,despite its strong theoretical advantages it is not commonly used. Further clinicalstudies in this patient subgroup are urgently needed.

INTRAOPERATIVE MANAGEMENT OF HEAD INJURIESAND OTHER CAUSES OF RAISED ICP

Trauma patients frequently have concomitant head injury. The principles ofanaesthesia for trauma patients with head injury are well established and coveredfully in the standard anaesthesia and neuroanaesthesia texts. A recently publishedtextbook of neuranaesthesia and critical care is listed in the suggestions for Furtherreading. Important principles worth emphasising include:

• Hyperventilation reduces ICP and brain volume and permits surgicalaccess to the brain. In an ideal world hyperventilation would be moni-tored by jugular venous oxygen content in view of the potential forischaemia if cerebral blood flow is reduced excessively.

• All volatile agents may increase cerebral blood flow and ICP. Hyper-ventilation is essential if volatile agents are used.

• Propofol infusions are increasingly popular.

• Ketamine may increase BP and ICP, and should be avoided.

• Full muscle relaxation is essential to avoid straining and coughinginduced increases in ICP.


88

Chap-06.qxd 2/2/02 12:59 PM Page 88

PRACTICAL CONDUCT OF ANAESTHESIA

• Conventional assessments of fitness for anaesthesia and surgery may notbe helpful. The bleeding patient may not be able to be stabilised untilthe bleeding is controlled.

• Many of these patients require ongoing resuscitation. The ABC systemis widely followed:

A is airway including cervical spine protection,B is breathing,C is circulation.

The less well-known system of VIP (Ventilation, Infusion and Perfusion)is perhaps more appropriate for surgical and trauma patients as itemphasises the interrelationship between ventilation and perfusion inoverall oxygen transport and because it reminds us that the cornerstoneof resuscitation in these patients is fluid infusion.

• It is an important principle that inotropes and vasopressors should not begiven as a substitute for fluids in the hypovolaemic patient but perfusionof the coronary and cerebral circulations must be maintained. It is, there-fore, appropriate to use such drugs to maintain perfusion of the heartand brain in the short term while one ‘catches up’ with blood loss.

• Critically ill patients do not always tolerate movement. Many willalready be intubated. Therefore, it seems logical to transfer thesepatients direct to theatre rather than via the anaesthetic room. In addi-tion, in many hospitals monitoring standards remain higher in theatrethan the anaesthetic room.

• Portable monitors with full invasive monitoring facilities are common-place and will be used for transfer of patients from ICU or A&E to the-atre. It may be sensible to continue to use this monitor rather than riskconfusion swapping over all the lines and cables. (Do not forget to plugin the portable monitor to maintain battery life for the journey back!)

• If invasive monitoring is not in situ it may be prudent (time permitting)to establish this using local anaesthesia prior to induction for beat-to-beat monitoring of this period of the anaesthetic (see below).

• Ruptured aneurysms and other cases of massive haemorrhage should be‘prepped’ on the table prior to induction as discussed in the chapter onvascular anaesthesia.

• Communication and timing with theatre staff, surgeons, porters, etc.should eliminate delays in potentially difficult circumstances and envi-ronments.


89

Chap-06.qxd 2/2/02 12:59 PM Page 89

INTRAOPERATIVE MONITORING

Full monitoring according to local and national protocols should be employed inall patients. In addition critically ill patients will require invasive monitoring withan indwelling arterial line for

• beat-to-beat monitoring of BP,

• sampling of blood for blood gas measurement,

• control of inotrope and vasopressor infusions;

and a central venous catheter for

• measurement of filling pressure, i.e. preload of the right ventricle (RV),

• guide to fluid requirements,

• infusion of irritant drugs, for example, inotropes, vasopressors and IVnutrition.

Central venous pressure (CVP) reflects right atrial pressure which is usually takento reflect RV end diastolic pressure. It does not necessarily reflect left ventricle (LV)preload and also poorly correlates with blood volume. CVP is often used as a guideto LV function. Directional changes in CVP may reflect alterations in LV perform-ance. However, if either ventricle becomes selectively depressed, or if there is severepulmonary disease, changes in CVP will not reflect changes in LV function.

Such patients may require a pulmonary artery flotation catheter (PAFC) to enablemeasurements of the filling pressures at the left side of the heart (estimated by thepulmonary capillary wedge pressure, the PCWP) as the inflated balloon at thecatheter tip is ‘wedged’ in the pulmonary artery and cardiac output and derivedhaemodynamic variables.

A urinary catheter is required for hourly urine volume measurement.Temperature should be monitored for all long procedures because of the dangersof perioperative hypothermia discussed above.

Monitoring strategies in the high risk surgical patient

• Invasive monitoring of elderly surgical patients has revealed a high inci-dence of ‘hidden’ abnormalities reflecting their reduced physiologicalreserve even in patients ‘cleared’ for surgery. Invasive monitoring dur-ing anaesthesia and in the postoperative period results in early recogni-tion of problems, ‘fine tuning’ of cardiovascular parameters and animproved outcome.13

• Perioperative optimisation (discussed elsewhere) of cardiac functionand oxygen transport will obviously require invasive monitoring ofcardiac function – most commonly with the aid of a PAFC.


90

Chap-06.qxd 2/2/02 12:59 PM Page 90

• Perioperative use of the PAFC is controversial. In the case of criticallyill patients there is doubt as to the value, in terms of improved outcome,of routine use of the PAFC with some studies even suggesting anincreased mortality associated with its use. Perioperative studies also cast doubt on the role of the PAFC in elective high risk surgery. Forexample, routine use of PAFCs during aortic surgery is not beneficial andmay lead to increased complications.14 Similarly there is no benefitfrom the PAFC for routine coronary artery bypass grafting (CABG)surgery.15

• The American Society of Anesthesiologists has published guidelines forits perioperative use16 and there is a large multicentre trial underway inthe UK which may resolve some of the controversy.

• Broad indications currently for the use of a PAFC are patients with severedisease of either ventricle, but most commonly patients with severe LVdysfunction, in order to optimise preload prior to the use of inotropes.

• In addition, the PAFC may enable early diagnosis of cardiac ischaemiaif there are sudden increases in PCWP, guide haemodynamic manage-ment of septic patients and monitor pulmonary artery pressures wherethese are elevated.

• Paradoxically, the PAFC is not always helpful in shocked or bleedingpatients in the operating theatre in whom the main aim of the anaes-thetist is often to administer sufficient fluids to enable the patient tosurvive the necessary ‘damage control’ surgery – followed by fine tuningof the haemodynamic state in the ICU.

FLUID THERAPY

The crystalloid versus colloid debate

• There has been controversy over the best type of fluid for resuscitation,i.e. crystalloids or colloids. Part of the problem is the lack of studiesshowing a sufficiently clear superiority of one fluid type over another,sufficient to convert its opponents and without reasonable criticisms ofstudy methodology. There are several problems with most of the avail-able studies, for example, different species, fluids, injuries, illnesses,complications studied.

• It is not widely appreciated that many of the original US studies ofcrystalloids versus colloids in trauma patients were flawed. This wasbecause most patients in both groups were given blood transfusions. Inthe US,patients are commonly given whole blood (as opposed to packedred cells in the UK), i.e. both groups received colloid from the whole


91

Chap-06.qxd 2/2/02 12:59 PM Page 91

blood, i.e. there was no such thing as a pure crystalloid group. Perhapsit is not surprising that few differences in outcome were detected.

• However, in most studies there is probably a skewed distribution ofseverity of sickness with a large group of patients who will do all rightwhichever fluid is given and a smaller group of patients who will dieregardless of which fluid is given. These patients may mask (statisticallyspeaking) a group of patients in whom choice of fluid may be critical.This possibility has been seized upon by the colloid enthusiasts!

There are certain statements regarding the colloid/crystalloid controversy whichcan be made which are reasonably accepted by both groups:

• Crystalloids replace interstitial losses. Colloids are superior at replacingplasma volume deficits – more quickly and lasting longer – giving greaterincreases in cardiac output and oxygen delivery. Crystalloid administrationmay also produce such increases but approximately three times as muchwill be needed with consequent delays in achieving goals of resuscitation.

• Crystalloids are cheap. Colloids are more expensive. Many centres inthe US use crystalloids almost exclusively. However, it has been pointedout that whole blood may be a significant source of colloid in studiespurporting to use no colloid.

• In most situations (for example routine surgery) both potentially giveexcellent results if appropriate amounts are used.

• Many studies show similar effects on respiratory function. Overdose ofeither may produce respiratory failure.

Most reasonable people do not take extreme positions in the debate. In most sit-uations close monitoring especially with regard to fluid overload is more impor-tant than absolute choice of fluid.

However, many believe in the ‘Golden Hour’ for resuscitation and that, therefore,speed of resuscitation is crucial. Therefore, when restoration of blood volume,cardiac output and tissue perfusion is urgent colloids are preferable to crystalloid.

Intraoperative volume loading

In the high risk or critically ill patient generous fluid loading may be appropriate:

• Fluid loading after induction of anaesthesia to a maximum stroke volumeled to a reduction of the incidence of low pHi, an index of gastricmucosal perfusion from 50% to 10%.17

• Intraoperative volume loading increases stroke volume and CO, result-ing in a more rapid postoperative recovery and a reduced hospital stayin fractured neck of femur patients.18


92

Chap-06.qxd 2/2/02 12:59 PM Page 92

This approach must be tempered with caution in the elderly or patients with knownheart failure due to the potential risk of fluid overload precipitating pulmonaryoedema. Perioperative invasive monitoring may be indicated.

However, inadequate fluid therapy is more common,more dangerous (organ hypo-perfusion leading to, for example, renal failure) and less easily treated than the effectsof excessive fluids which can, if necessary, be cleared with diuretic therapy.

Permissive hypovolaemia

• An important study in 1994 of penetrating trauma showed an improvedsurvival in those patients with ‘delayed’ fluid resuscitation, i.e. minimalIV fluids given prior to definitive operative intervention.19 This hasbeen called ‘permissive hypovolaemia’.

• The rationale, borne out by previous animal studies, is that full resusci-tation results in

– higher BP disrupting clot formation,

– haemodilution and decreased viscosity disrupt clot formation,

– dilutional coagulopathy.

• The recommendation has, therefore, been made in penetrating injuryto limit fluids to maintain an MAP not � 50 until bleeding has beensurgically controlled, then full resuscitation.

• The biggest problem is that this study was performed in penetratinginjuries. Patients with blunt trauma (the majority) are not so likely tohave definitive surgical interventions.

• This approach is also applicable in surgical cases of life-threateninghaemorrhage, for example, ruptured aortic aneurysm.

Blood transfusion

From the perspective of the anaesthetist certain points are worth emphasising:

• The importance of communication with surgeon re-bleeding. Onoccasion the surgeon may need to be told to stop dissecting and con-trol active bleeding to allow one to ‘catch up’.

• Similarly one must communicate early with the blood bank re-requirements especially requirements for clotting factors.

• Many anaesthetists only start to consider blood transfusion once approxi-mately 10% of the patient’s blood volume (based on 80 ml/kg bodyweight) has been lost. With ongoing brisk haemorrhage one should notwait until 10% has been lost!


93

Chap-06.qxd 2/2/02 12:59 PM Page 93

• Maintaining body temperature will minimise coagulopathy and bloodloss as previously described.

• Maintaining blood volume is probably more important in the shortterm than maintaining Hb. However, with major haemorrhage bloodwill be needed!

• Autologous transfusion systems, for example, ‘cell savers’ should beconsidered for appropriate ‘clean’ operations.

INOTROPES AND VASOACTIVE DRUGS

In addition to the normal anaesthetic goals that pertain to all patients one must pay special interest to the maintenance of organ blood flow and function in the critically ill patient. This is obviously the case for all our patients but fortunatelythe vast majority of low risk patients present few problems and rarely need anyform of circulatory support. In septic or shocked patients this is the norm and the choice of inotropes and vasopressors and monitoring of the circulation are discussed below:

• Adequate filling pressures and intravascular volume are crucial prior toanaesthesia and also the use of inotropic agents. With hypovolaemia thevasodilator effects of inotropic agents such as Dobutamine predominateleading to hypotension. The use of vasopressors in hypovolaemia willreduce splanchnic and muscle blood flow. Indeed, the use of noradren-aline in haemorrhagic shock is a useful animal model of acute tubularnecrosis!

• Cardiac function can be severely compromised in haemorrhagic shockso that an element of cardiogenic shock contributes to the shocked state.In such cases the response to resuscitation may be compromised andinvasive monitoring and/or inotropes required as detailed below. As earlyas the 1950s the contribution of the heart to progressive, irreversibleshock was recognised and it was also demonstrated that the homeostaticmechanisms and vasoconstriction were not sufficient to maintain cor-onary perfusion in severe haemorrhage. Therefore, cardiac dysfunctionneeds to be detected and corrected as early after injury as possible.

• For myocardial support in the failing heart and low output statesDobutamine is probably the agent of choice.

• For vascular support, for example, with abnormal vasodilation, nor-adrenaline is probably the agent of choice.

• In view of concerns relating to gut blood flow and lactic acidosis therole of adrenaline infusions perioperatively is controversial.


94

Chap-06.qxd 2/2/02 12:59 PM Page 94

• For information on the support of regional circulations including theuse of Dopamine and Dopexamine the chapters on PerioperativeOptimisation and Renal Insufficiency should be consulted.

Many critically ill patients in the operating theatre will have severe sepsis. The latest international guidelines on management of septic shock support the use ofnoradrenaline as first line agent in septic shock once volume resuscitation has beenachieved.20 The two main arguments against the use of noradrenaline, i.e.

• reduced renal blood flow,

• reduced cardiac output secondary to vasoconstriction induced increasesin afterload

are unfounded. In fact, in septic shock noradrenaline improves urine volumes andcreatinine clearance21 and improves cardiac output.22

The possibility of altered organ blood flow intraoperatively due to interactionsbetween anaesthetic agents and vasoactive drugs is poorly understood.

OXYGEN TRANSPORT IN THE HIGH RISK OR CRITICALLYILL SURGICAL PATIENT

Differences between haemorrhagic shock and traumatic shock

Haemorrhage results in well-known physiologic changes. Traumatic shock includesthese responses but they are modified by the tissue injury and its associated inflam-matory response. This has several practical effects:

• HR responds to haemorrhage by an initial tachycardia followed even-tually by a progressive bradycardia. This has been labelled as a ‘paradox-ical bradycardia’ but there is nothing paradoxical about the heartslowing in the absence of adequate venous return in an attempt tomaintain stroke volume. This is seen following, for example, rupturedectopic pregnancy. With tissue injury there is no late slowing of theheart and tachycardia continues.

• BP is maintained by vasoconstriction until more than one-third of blood volume has been lost. With tissue injury, BP is maintained to agreater degree by the surge in catecholamines and other nociceptivestimuli but this is at the expense of tissue perfusion due to excessive vasoconstriction.

• Animal studies show that for an equivalent degree of blood loss,traumatic injury results in greater tissue hypoperfusion and a greater‘injury’ than simple haemorrhage.


95

Chap-06.qxd 2/2/02 12:59 PM Page 95

• The wound and fracture sites are metabolically active with a resultantrequirement for increased oxygen consumption and glucose oxidation –the concept of ‘the wound as an organ’. In addition to the local reasonsfor increased metabolic demands, there are systemic inflammatory andcatabolic causes of increased metabolic demand requiring an increasedcardiac output compared to normal. This may imply a need for increasedcardiac output and oxygen delivery in trauma and high risk surgicalpatients – Shoemaker’s optimal goals as discussed in the chapter onPerioperative Optimisation.

Optimal goals

The best evidence for a beneficial therapeutic effect of maximising oxygen trans-port according to previously identified optimal goals is from studies where therapywas initiated very early in the presence of tissue hypoperfusion, i.e. preoperativelyin high risk surgical patients. Shoemaker’s original prospective, randomised studydemonstrating the virtues of optimising oxygen transport was performed in surgicalpatients.23

A thought provoking review24 points out that there are conflicting priorities in managing surgical patients at risk of myocardial ischaemia, for example, using � blockers and those in whom the cardiac output and oxygen delivery need to be increased, for example, using inotropes. Both groups are at risk of an adverseoutcome but the approach is different and identification of the group that patientsbelong to is important. These issues are further explored in the chapter onPerioperative Optimisation.

Oxygen debt and lactic acidosis in the high risk surgical patient

Even when oxygen delivery is well maintained oxygen consumption falls underanaesthesia. Animal studies show that anaesthesia reduces tissue oxygen extractionespecially in septic models. This occurs with all agents but is associated with lac-tic acidosis only with volatile agents. Ketamine may increase oxygen extraction bythe tissues compared with volatile and IV barbiturate techniques.25

Thus, although anaesthesia reduces metabolic rate and oxygen demand this maybe countered in the critically ill patient by the reduction in the tissue’s ability toextract oxygen. An oxygen debt may develop especially if there are falls in cardiacoutput and/or oxygen delivery below a critical level.26 Worryingly, the reductionof tissue oxygen extraction under anaesthesia may increase the threshold for oxygendelivery to be ‘critical’,27 i.e. lesser degrees of fall in cardiac output and oxygendelivery may result in tissue hypoxia under anaesthesia.

This has obvious implications for anaesthesia of the critically ill or shockedpatient, in whom maintenance of cardiac output and oxygen delivery are crucial.


96

Chap-06.qxd 2/2/02 12:59 PM Page 96

There are many studies demonstrating that high risk surgical patients develop anintraoperative ‘oxygen debt’, the magnitude and duration of which correlates withthe development of lactic acidosis, organ failure and increased mortality.28–30 Thisoxygen debt is postulated to potentially arise from anaesthetic cardiac depression,direct anaesthetic reductions in tissue oxygen uptake as already described, failure tomaintain adequate fluid intake during surgery and perhaps hypothermia.

The crucial message is that high risk surgical patients may have reduced cardiacreserves, especially in the elderly, suffer occult tissue hypoperfusion with a develop-ing oxygen debt postoperatively, proceed to multiple organ failure if there is nointervention to reverse the tissue hypoperfusion and have a higher mortality thanpatients who do have sufficient reserves to reverse their oxygen debt and preventserious tissue hypoxia. Appropriate interventions may include fluid therapy, oxygen,inotropes and vasopressors.

However, recent studies suggest that not all metabolic acidosis under anaesthesia isdue to oxygen debt and/or lactic acidosis. Rapid infusion of 0.9% saline can causesignificant hyperchloraemic metabolic acidosis.31,32 The common treatment ofadministering more fluid for intraoperative acidosis may be inappropriate if thefluid administered has a high chloride content such as saline or Gelofusine.Measurement of serum lactate and chloride may be helpful in distinguishing thecause of the intraoperative acidosis.

Further reading

Intensive Care Society. Guidelines for Transport of the Critically Ill Adult, 1997.

The Association of Anaesthetists of Great Britain and Ireland and TheNeuroanaesthesia Society of Great Britain and Ireland. Recommendations for theTransfer of Patients with Acute Head Injuries to Neurosurgical Units, 1996.

Grande CM. Textbook of Trauma Anesthesia and Critical Care, 1993. St Louis: Mosby.

Matta BF, Menon DK, Turner JM (eds). Textbook of Neuroanaesthesia and CriticalCare, 2000. London: Greenwich Medical Media.

References

1. Schmied H, Kurz A, Sessler DI, Kozek S, Reiter A. Mild hypothermiaincreases blood loss and transfusion requirements during total hip arthroplasty.Lancet 1996; 347: 289–92.

2. Frank SM, Fleisher LA, Breslow MJ, Higgins MS, Olson KF et al.Perioperative maintenance of normothermia reduces the incidence of morbidcardiac events: a randomised clinical trial. JAMA 1997; 227: 1127–43.


97

Chap-06.qxd 2/2/02 12:59 PM Page 97

3. Durieux ME, Sperry RJ, Longnecker DE. Effects of hypoxemia on regionalblood flows during anesthesia with halothane, enflurane, or isoflurane.Anesthesiology 1992; 76: 402–8.

4. Mayer N, Zimpler M, Kotai E et al. Enflurane alters compensatory hemo-dynamic and humoral responses to hemorrhage. Circ Shock 1990; 30: 165–70.

5. Bogetz MS, Katz JA. Recall of surgery for major trauma. Anesthesiology 1984;61: 6–9.

6. Longnecker DE, Sturgill BC. Influence of anesthetic agent on survival follow-ing hemorrhage. Anaesthesiology 1976; 45: 516–21.

7. Idvall J. Influence of ketamine anesthesia on cardiac output and tissue perfu-sion in rats subjected to hemorrhage. Anesthesiology 1981; 55: 297–304.

8. Pedersen T, Engback J, Klausen NO et al. Effects of low dose ketamine andthiopentone on cardiac performance and myocardial oxygen balance in highrisk patients. Acta Anaesthesiol Scand 1982; 26: 235–9.

9. Lippmann M, Appel PL, Mok MS et al. Sequential cardiorespiratory patternsof anesthetic induction with ketamine in critically ill patients. Crit Care Med1983; 11: 730–4.

10. Yli-Hankala A, Kirvela M, Randell T et al. Ketamine anaesthesia in a patientwith septic shock. Acta Anaesthesiol Scand 1992; 36: 483–5.

11. Van der Linden P, Gilbart E, Engelman E et al. Comparison of halothane,isoflurane, alfentanil and ketamine in experimental septic shock. Anesth Analg1990; 70: 608–17.

12. Nader-Djalal N,Knight PR,Bacon MF et al. Alterations in the course of acid-induced lung injury in rats after general anesthesia: volatile anesthetics versusketamine. Anesth Analg 1998; 86: 141–6.

13. Del Guercio LRN, Cohn JD. Monitoring operative risk in the elderly. JAMA1980; 297: 845–50.

14. Valentine RJ, Duke ML, Inman MH et al. Effectiveness of pulmonary arterycatheters in aortic surgery: a randomized trial. J Vasc Surg 1998; 27: 203–11.

15. Tuman KJ,McCarthy RJ, Spiess BD et al. Effect of pulmonary artery catheter-ization on outcome in patients undergoing coronary artery surgery.Anesthesiology 1989; 70: 199–206.

16. Practice guidelines for pulmonary artery catheterization. A report by theAmerican Society of Anesthesiologists Task Force on Pulmonary ArteryCatheterization. Anesthesiology 1993; 78: 380–94.


98

Chap-06.qxd 2/2/02 12:59 PM Page 98

17. Mythen MG, Webb AR. Perioperative plasma volume expansion reduces theincidence of gut mucosal hypoperfusion during cardiac surgery. Arch Surg1995; 130: 423–9.

18. Sinclair S, James S, Singer M. Intraoperative intravascular volume optimisationand length of hospital stay after repair of proximal femoral fracture: randomisedcontrolled trial. Br Med J 1997; 315: 909–12.

19. Bickell WH, Wall MJ Jr, Pepe PE, Martin RR, Ginger VF et al. Immediateversus delayed fluid resuscitation for hypotensive patients with penetratingtorso injuries. N Engl J Med 1994; 331: 1105–9.

20. Vincent JL. Hemodynamic support in septic shock. Inten Care Med 2001; 27:S80–92.

21. Redl-Wenzl EM,Armbruster C, Edelmann G, Fischl E, Kolacny M et al. Theeffects of norepinephrine on hemodynamics and renal function in severe septic shock states. Inten Care Med 1993; 19: 151–4.

22. Martin C, Viviand X, Arnaud S et al. Effects of norepinephrine plus dobuta-mine or norepinephrine alone on left ventricular performance of septic shockpatients. Crit Care Med 1999; 7: 1708–13.

23. Shoemaker WC,Appel PL, Kram HB,Waxman K, Lee TS. Prospective trial ofsupranormal values of survivors as therapeutic goals in high-risk surgicalpatients. Chest 1988; 94: 1176–86.

24. Juste RN,Lawson AD,Soni N. Minimising cardiac anaesthetic risk: the tortoiseor the hare? Anaesthesia 1996; 51: 255–62.

25. Van der Linden P, Gilbart E, Engelman E et al. Effects of anesthetic agents onsystemic critical O2 delivery. J Appl Physiol 1991; 71: 83–93.

26. Lugo G, Arizpe D, Dominguez G. Relationship between oxygen consump-tion and oxygen delivery during anesthesia in high-risk surgical patients. CritCare Med 1993; 21: 64–9.

27. Van der Linden P, Schmartz D, Gilbart E et al. Effects of propofol, etomidate,and pentobarbital on critical oxygen delivery. Crit Care Med 2000; 28: 2492–9.

28. Shoemaker WC, Appel PL, Kram HB. Role of oxygen debt in the develop-ment of organ failure sepsis, and death in high-risk surgical patients. Chest1992; 102: 208–15.

29. Shoemaker WC,Appel PL, Kram HB. Tissue oxygen debt as a determinant oflethal and nonlethal postoperative organ failure. Crit Care Med 1988; 16:1118–20.


99

Chap-06.qxd 2/2/02 12:59 PM Page 99

30. Hess W, Frank C, Hornburg B. Prolonged oxygen debt after abdominal aorticsurgery. J Cardiothorac Vasc Anesth 1997; 11: 149–54.

31. Scheingraber S, Rehm M, Sehmisch C et al. Rapid saline infusion produceshyperchloremic acidosis in patients undergoing gynecologic surgery.Anesthesiology 1999; 90: 1265–70.

32. Waters JH, Miller LR, Clack S et al. Cause of metabolic acidosis in prolongedsurgery. Crit Care Med 1999; 27: 2142–6.


100

Chap-06.qxd 2/2/02 12:59 PM Page 100

101

7THE ELDERLY PATIENT

Data from the Office for National Statistics1 showed that, in 1997, the average life expectancy for a man was 74 years and for a woman 79 years. The mid-1999population demographics revealed that 9.8 million (16.5%) of the population of the United Kingdom were over pensionable age and that was expected to riseby another 4.6 million by the year 2023:

• With the advancement of anaesthetic and surgical techniques,more andmore elderly patients are presenting for major elective and emergencysurgery.

• It is vital therefore that the practising anaesthetist is aware of the important differences that exist between the elderly patient and theyoung adult.

Ageing is a continuous process once the organism has reached maturity. There is no strict, defined age when an adult becomes elderly. In this chapter, like othertexts, the elderly patient will be assumed to be aged 65 years or over.

PHYSIOLOGICAL CHANGES ASSOCIATED WITH AGEING

After the age of 30 years there is a gradual deterioration in organ function. Therate and extent of decline often determines those who are ‘physiologically youngfor their age’ or those who are ‘physiologically old for their age’.

The ageing cardiovascular system2–6

Most of the investigation of the cardiovascular system in human adults comes fromlongitudinal studies of cohorts of adults as they age and in aged individuals with no heart disease. Most investigation has been with echocardiography or angiog-raphic or radionuclide imaging of the heart. Whether the changes in the vascularsystem lead to compensatory changes in the heart or whether both occur

Chap-07.qxd 2/2/02 12:59 PM Page 101

simultaneously and independently is a matter of debate:

• The arterial system becomes less compliant due to a loss in elastic tissuein the vessel wall. This results in an increased left ventricular afterloadand systolic hypertension. The arteries also become less responsive tovasodilators such as nitric oxide, atrial naturetic peptide and �2 adreno-ceptor stimulation.

• The venous system also becomes less compliant with a reduction in the strength of smooth muscle contraction within the vessel wall. The elderly therefore have less blood in the capacitance vessels and less ability to squeeze this blood into the central circulation in the face ofintravascular fluid depletion.

• The ventricle hypertrophies with age. This may be in part as a responseto the increased afterload and as a primary effect of ageing. Ventricularhypertrophy reduces ventricular compliance, increases left ventricularend diastolic pressure (LVEDP) and reduces early diastolic filling of theventricle. The elevated LVEDP increases the importance of atrialcontraction (hence sinus rhythm) on late ventricular filling. Atrial hyper-trophy develops to the increased impedance (LVEDP) to atrial emptying.

• The myocardium and pacemaker cells become less responsive to �2 adrenoceptor stimulation. Therefore there is a reduction in bothinotropic and chronotropic effects of �2 stimulation.

• At rest cardiac index is unchanged or reduced in proportion to thereduction in basal metabolic rate or silent coronary artery disease.The situation during exercise is markedly different to the young adult.In the exercising young adult, cardiac output is increased by an increasedheart rate and ejection fraction (i.e. a lower left ventricular end diastolicand systolic volume (LVEDV and LVESV)). In the elderly, heart ratefalls during exercise, LVEDV increases (by 20–30%) but LVESVdecreases less, and therefore ejection fraction increases less, than in theyoung adult. It is apparent then, that cardiac output in the elderlypatient is more pre-load dependent than in the young adult duringtimes of cardiovascular stress.

• Pacemaker activity of the heart declines with age. The cells of the sino-atrial (SA) node atrophy, conduction through the atrioventricular (AV)node is increased and conduction through the bundles is impaired.Heart block, bundle branch block and arrhythmias (both brady- andtachyarrythmias) become increasingly common with age.

• Coronary artery vascular resistance increases in the elderly because of the increased LVEDP and ventricular hypertrophy, but the reduced


102

Chap-07.qxd 2/2/02 12:59 PM Page 102

coronary flow is counterbalanced by a reduced myocardial oxygenconsumption.

Ageing of the respiratory system7–9

As one ages there are changes in the structure of the lung and airways along withchanges in the thoracic wall. These fundamental structural changes lead to thephysiological changes seen with advancing age:

• There is a loss in elastic tissue within the lung parenchyma as well as loss of alveolar surface area and therefore loss in surface tension forces.Both elastic tissue and surface tension contribute to the elastic recoil of the lung, hence the compliance of the ageing lung is increased(compliance being the reciprocal of elastance). Calcification of thecostal cartilage and the rib articulations reduce the thoracic compliancethat counterbalances the increased lung compliance. There is somedebate as to whether total compliance is unaltered or reduced becauseof the greater reduction in thoracic compliance over the increase inlung compliance.

• The losses in alveolar surface area results in V/Q mismatch, an increasedphysiological shunt (increased A-a gradient) and consequently a lowerPaO2. The PaO2 can be estimated from the formula: PaO2 (mmHg) �100 � Age/3.

• Changes in lung volumes also contribute to an increased physiologicalshunt. Throughout life, there is an increase in the volume of airrequired to prevent small airway collapse also known as closing volume(CV). At around 45 years of age, CV exceeds functional residual capac-ity (FRC) in the supine position and in the seated position by 65 yearsof age. Once CV exceeds FRC then airway closure occurs during tidalventilation. The increase in CV can, on the whole, be explained by theloss in elastic tissue with age.

• Aside from an increase in CV with age there is an increase in residualvolume. FRC, the point at which the outward pull of the thorax isbalanced by the tendency for the lung to collapse, is unchanged at theexpense of a reduced expiratory reserve volume (ERV). As ERV isreduced it follows that vital capacity (VC) must be reduced. It is believedthat total lung capacity is unchanged, or only reduced slightly (10%)with age.

• The large airways increase in size as one ages resulting in an increasedanatomical and physiological deadspace. Airway resistance is unchangedas the resistance (proximal) airways dilate and the smaller, distal, airways

THE ELDERLY PATIENT

103

Chap-07.qxd 2/2/02 12:59 PM Page 103

collapse thus offsetting each other. Although total compliance isunchanged or marginally reduced, the loss in elasticity of the lungs andrigidity of the chest wall increases the work of breathing.

• The elderly have a diminished response to both hypercapnia and hypoxia.The elderly, like the younger adult are able to increase respiratory ratebut are unable to increase tidal volume in response to an abnormalPaO2 or PaCO2. The reason for the fall in tidal volume is postulated tobe a reduced sensitivity or a reduced output from the respiratory centrerather than a loss in respiratory muscle power with age.

• The elderly have blunted protective laryngeal reflexes and therefore aremore at risk of pulmonary aspiration during anaesthesia.

• Pulmonary vascular resistance increases with age but it is doubtful if thisis of any clinical significance.

Changes in renal function with age10,11

Data regarding the changes in the kidney with age is primarily from cross-sectionalstudies and histological findings. Some data is available from longitudinal studiesand tends to be more reliable than the former sources because it excludes renaldysfunction as a result of age related changes.

• Renal mass declines with age. After the 3rd decade there is 1% loss peryear. The reduction in mass is due to glomerular loss (up to 30% by the 8th decade) which is predominantly cortical. The exact cause of the glomerular atrophy is unknown but it mirrors a reduction in renalblood and plasma flow (10% per decade).

• Loss of glomeruli has been implicated in the fall in glomerular filtrationrate (GFR) with age. Absolute creatinine clearance falls approximately1 ml/min/1.73 m2 per year, or from 140 ml/min/1.73 m2 in the 3rddecade to 97 ml/min/1.73 m2 in the 8th decade. However, plasmacreatinine levels are unchanged in the elderly because a reduced musclemass results in a reduced production of creatinine.

• Renal tubular function declines with age. Inulin clearance, which rep-resents tubular secretory function declines and is paralleled by deterior-ation in reabsorptive function. Tubular dysfunction may be explainedby the loss of glomerular units and a reduction in metabolically activetubular cells with age.

• The aged kidney is less effective at concentrating urine and conservingwater in the face of water deprivation. This may result from a loweringin the medullary concentration gradient caused by a disturbance of the


104

Chap-07.qxd 2/2/02 12:59 PM Page 104

counter-current mechanism due to alterations in renal blood flow anda relative resistance to anti-diuretic hormone (ADH). Moreover thirstperception during periods of dehydration is impaired. The nephron is also impaired in its ability to dilute the urine in the face of wateroverload.

• The elderly face problems in salt conservation. Plasma renin and aldosterone levels are reduced in the elderly. This may be due to the relative unresponsiveness to �2 receptor stimulation as renin is releasedin response to �2 adrenoceptor stimulation. Moreover, changes in theheart with age lead to atrial distension and release of atrial natriuretic factor (ANF) which also suppresses renin and aldosterone release. Notonly does the relative deficiency of these two hormones lead to sodiumloss but it places the elderly at risk of hyperkalaemia.

The effect of age on hepatic function12,13

The liver, like most other organs, involutes with age, so by the 8th decade the liverhas lost two-fifths of its mass. There is also a reduction in hepatic blood flow thatnot only reflects the loss in hepatic cellular mass but also an absolute reduction interms of percentage of cardiac output. Despite the reduction in mass and bloodflow, it appears that hepatocellular enzyme function is preserved with advancingage. In vitro studies in patients with normal histology on liver biopsy failed todemonstrate any deterioration in hepatic microsomal oxygenase or hydrolaseactivity (phase I metabolic reactions) and also showed that reduced glutathione(phase II conjugation reactions and a major hepatic anti-oxidant) concentrationsare maintained.

In parallel with the apparent preservation of hepatocellular function, serum con-centration of bilirubin, alkaline phosphatase, and transaminases are unaffected byage. Coagulation studies are also unchanged by age but there is a gradual declinein serum albumin concentration.

Changes in the nervous system with age14,15

Memory loss, confusion and dementia are the clinical manifestations of ageing of the brain. Unlike other organs there are no readily applicable tests of ‘brainfunction’ but the following are generally accepted as age related changes, with orwithout clinical manifestation:

• Normal pressure hydrocephalus results from global atrophy of the brainand an increase in cerebrospinal fluid (CSF) volume. The brain weighs20% less by the 8th decade than in the 2nd decade of life and CSF volume increases by 10% in the same time period.

THE ELDERLY PATIENT

105

Chap-07.qxd 2/2/02 12:59 PM Page 105

• Cerebral blood flow is reduced in line with brain volume but auto-regulation to carbon dioxide and mean arterial blood pressure is preserved.

• Within the brain the most metabolically active cells (grey matter of thecerebral and cerebellar cortices, basal ganglia, thalamus) atrophy morethan the white matter. Regional blood flow reflects the neuronal losswith flow to the grey matter reduced more than that to the white.

• The levels of excitatory neurotransmitters (norepinephrine, serotonin,dopamine and tryrosine) are reduced.

The peripheral neurones like their counterparts in the brain undergo age relateddegeneration. In particular there is:

• An increased threshold to stimulate sensory organs, such as pain cor-puscles, and a reduced conduction velocity in afferent neurones andascending spinocortical tracts. There is also a reduced conduction vel-ocity in motor neurones and in the corticospinal tracts so that the reflexarc for painful stimuli is increased and righting reflexes are impaired.

• Skeletal muscle mass is reduced and extrajunctional acetylcholine recep-tors increase in response to degeneration of motor neurones.

Neuroendocrine changes with age16

Ageing produces a state akin to a hyperadrenergic state. The impaired responses inthe elderly to �2 adrenoceptor stimulation leads to increased plasma norepinephrineand epinephrine concentrations (2–4-fold) despite atrophy of the adrenal medulla.Cardiovascular reflexes are also impaired in the elderly. Reduced responsiveness of the baroreceptors results in an underdamped cardiovascular system and there isa reduced vasoconstrictor response to cold with less heart rate change in responseto changes in posture. The elderly are therefore more vulnerable to cardiovascularinstability, particularly during sympathetic blockade.

Changes in body fluid composition and metabolism with ageing

The key changes that occur are summarised below:

• Basal metabolic rate falls as a consequence of a reduced skeletal massand a reduction in the metabolically active areas of the brain, kidneyand liver.

• Increased body fat results in a reduction in total body water.

• Testosterone and tri-iodothyronine levels are reduced.

• Glucose intolerance occurs.


106

Chap-07.qxd 2/2/02 12:59 PM Page 106

CHANGES IN PHARMACOKINETICS ANDPHARMACODYNAMICS WITH AGE17 ,18

In general absorption of drugs from the gastrointestinal tract is unaffected by age.There are, however, important changes in distribution, metabolism and elimin-ation of drugs because of age related changes of the organs.

• A reduction in total body water means that the volume of distribution ofwater soluble drugs (e.g. non-depolarising muscle relaxants) is decreasedwith an effective increase in the tissue concentration. Conversely, anincrease in body fat results in an increased volume of distribution forlipid soluble drugs.

• The reduction in albumin concentration in the elderly increases thefree fraction of protein bound (i.e. lipid soluble) drugs and thereforeincreases the bioavailabilty at their effector sites.

• Hepatic clearance of a drug is dependent on three factors, the intrinsicclearance (CLint), the free fraction of the drug ( f ) and the hepatic bloodflow (QH). The hepatic clearance of drugs with a low CLint is depend-ent on CLint and f and are said to be ‘capacity limited’. Examples ofsuch drugs are barbiturates, benzodiazepines and theophyllines. If thefree fraction of a highly protein bound drug is increased, then the hepaticclearance becomes more dependent upon QH than CLint. The elderlyhave a reduced QH but CLint is largely unchanged. Therefore thehepatic clearance of capacity limited drugs with low protein binding isunchanged with age. The reduction in serum albumin will increase f ofhighly protein bound drugs (e.g. thiopentone) and so their hepaticclearance will be reduced as a result of a reduced QH.

• Drugs with a high CLint will be dependent on QH only for the hepaticclearance. They are said to be ‘flow limited’ and their clearance will be reduced as a result in the age related fall in QH. Examples of flowlimited drugs are �-blockers, tricyclic anti-depressants, opioid anal-gesics and amide local anaesthetics.

• Biliary excretion of drug metabolites is unaffected by age, but renalexcretion of water soluble drugs and drug metabolites may be reducedby age related reduction in GFR and tubular secretion.

• As well as changes in drug pharmacokinetics (e.g. increased free frac-tion of drugs, reduced volume of distribution, reduced clearance) theincreased sensitivity to some drugs in the elderly is also due to pharma-codynamic changes. The reduction in excitatory neurotransmitters in the brain with grey matter atrophy is thought to be the basis for the enhanced sensitivity to intravenous induction agents and reduced

THE ELDERLY PATIENT

107

Chap-07.qxd 2/2/02 12:59 PM Page 107

MAC to volatile anaesthetics. Changes in receptor sensitivity may alsoaccount for the enhanced analgesia seen with morphine, and alteredsensitivity to benzodiazepines.

CO-EXISTING DISEASE AND AGE RELATED ORGAN DYSFUNCTION

The deterioration in the various organ systems described above can be acceleratedand worsened by co-existing disease. These diseases are more likely to be encoun-tered with advancing age:

• Hypertension, (essential or secondary to other diseases), diabetes mellitus,smoking and hyperlipidaemia all predispose to atheromatous disease ofthe arteries. This may present as angina or myocardial infarction, cere-brovascular disease, peripheral vascular insufficiency and abdominalaneurysm formation.

• Cardiac function may also be worsened by valvular abnormalities.Rheumatic fever, age related fibrosis and calcification can lead to stenoticvalves, whilst ischaemic heart disease, rheumatoid arthritis (RA), connec-tive tissue diseases (CTD), hypertension and even stenotic valves (aortic)may result in regurgitant valves.

• Pulmonary function is particularly affected by smoking and can resultin emphysema or chronic bronchitis. Chronic asthma may also lead tofixed obstructive airways disease.

• Glomerulonephritis, hypertension, diabetes mellitus, RA, CTD andatheroma of the abdominal aorta and/or renal arteries can cause pre-mature renal failure. It should be remembered that renal failure is animportant cause of hypertension.

• Chronic alcohol ingestion is the major cause of cirrhosis and hepato-cellular failure and may be associated with a dilated cardiomyopathy.Other rarer causes of liver dysfunction are primary biliary cirrhosis,chronic active hepatitis (post viral or autoimmune), �1 antitrypsin defi-ciency (associated with emphysema) and drug therapy.

• It is important not to forget that drug therapy for medical conditionsmay adversely affect some organs. Examples would include renal dam-age from use of non-steroidal anti-inflammatory agents and penicil-lamine used in the treatment of RA. The liver particularly can beadversely affected by a long list of drugs and this should be borne inmind if faced with abnormal liver function tests or jaundice.

• Acute confusional states in the elderly may also be drug induced and usually resolve once the drug is discontinued.


108

Chap-07.qxd 2/2/02 12:59 PM Page 108

ANAESTHESIA FOR THE ELDERLY PATIENT

The two recent CEPOD reports19,20 highlighted the impact that the elderly haveupon anaesthetic and surgical specialties:

• The 2000 report showed that the number of elderly patients (over 60years for females and over 65 for males) presenting for surgery hadincreased from the 1990 report.

• In 1998/99, over 90% of patients were aged 60 years or more, with 38%over 80 years of age.

• Only 35% of procedures were deemed to be elective or scheduled,whilst 50% were urgent and the remainder emergency procedures.

• The majority of the elderly patients presented for general (42%),orthopaedic (22%) or vascular procedures (14%) and 84% were deemedby the anaesthetist to be of ASA 3 or more.

One of the key points in the 2000 report was that:

• ‘The profile of patients who die within 30 days of an operation haschanged since the report of 1990. Patients are more likely to be older,have undergone an urgent operation, be of poorer physical status andhave co-existing cardiovascular or neurological disorder’.

The 1999 CEPOD report that looked specifically at patients over 90 years at the time of operation recognised that ‘elderly patients have a high incidence of co-existing disorders and a high risk of early post-operative death’.

Pre-operative preparation

The pre-operative visit is essential for:

• initiating the patient – anaesthetist relationship and helping allay anxiety,

• determining the presence of co-existing diseases,

• planning any pre-operative investigations,

• choice of anaesthetic technique,

• method of post-operative analgesia,

• determining post-operative placement (ward, high depency unit (HDU),intensive care unit (ICU)).

The pre-operative visit for the elderly is often more taxing and takes longer thanin the younger adult. Elderly patients may have cognitive impairment, memoryloss and impaired hearing and vision. Moreover they might not understand what

THE ELDERLY PATIENT

109

Chap-07.qxd 2/2/02 12:59 PM Page 109

an anaesthetist is or does! Extraction of information can be prolonged and diffi-cult, so it is vital that the patient’s notes be available for perusal.

The elderly patient should have the same assessment as a younger patient, but withparticular emphasis on

• A functional assessment of their cardiorespiratory status. It is importantto realise that the elderly often have different symptoms of a disease.For example, ischaemic heart disease will often present as dyspnoearather than chest pain. The reason can be explained on the basis of the age related cardiac changes, in that myocardial ischaemia furtherelevates the LVEDP and results in pulmonary oedema. In general a person able to climb a flight of stairs or walk up a gentle hill has a lowerpost-operative cardiac mortality than one who is housebound by theirsymptoms.

• Assessment of hydration is important but also difficult. As emphasisedbefore, the elderly are prone to dehydration during times of fasting andhypovolaemia worsens cardiac performance and increases post-operativecomplications. The signs of dehydration such as loss of skin turgor, dryeyes and mouth are common findings in the elderly so one will have tolook for more subtle signs such as loss of jugular venous pulsation in thesupine position, postural hypotension and a raised urea. Fluid balancecharts should be consulted to help with assessment of fluid status.Dehydration should be corrected pre-operatively with the use of centralvenous pressure monitoring as necessary to prevent tipping the patientinto pulmonary oedema.

• The presence of cardiac murmurs, particularly of the aortic valve,should be sought, especially if a regional anaesthetic technique is beingconsidered.

• The history and examination of the patient largely determines pre-operative investigations. It is generally agreed that all patients over 65years of age should have a full blood count, urea and electrolytes and anECG. One must realise that these investigations may show no abnor-mality despite the presence of age related organ dysfunction. Whenordering more advanced investigations one should give thought to the accuracy of the results. For example, an exercise ECG may be of limited value when the patient is disabled by arthritis so radio-nuclide imaging or stress echocardiography of the heart may be moreappropriate.

• Elderly patients should not be denied premedication but the drugs pres-cribed should be done so with knowledge of the altered pharmaco-kinetics and dynamics in the elderly.


110

Chap-07.qxd 2/2/02 12:59 PM Page 110

• The age, physical status of the patient, the degree of urgency and the type of surgery performed determine post-operative outcome.Therefore very careful consideration should be given to the risk–benefitwhen an elderly patient of poor physical status presents for majorsurgery. Where risks outweigh perceived benefit then surgery shouldbe deferred.

Anaesthetic technique

The overriding goal of the anaesthetist is to maintain global oxygen delivery to thepatient. In effect this means:

• avoidance of hypotension/hypovolaemia,

• avoidance of hypoxia,

• avoidance of anaemia,

• these principles must be adhered to in elderly patients along with theavoidance of hypothermia.

Specific problems that can be encountered during anaesthesia for the elderly are:

• The elderly often have fragile veins making venous access difficult.

• Elderly patients have thin skin and arthritic joints. Special care shouldbe taken when transferring and positioning on the operating table. Allbony prominences should be well padded.

• Elderly patients are more at risk of hypothermia both during generalanaesthesia (GA) and regional anaesthesia.21,22 Warming mattresses,warmed intravenous fluids and warm air blowers must be readily avail-able and used for all but the shortest of cases.

• The 1999 CEPOD report19 highlighted the high incidence of intra-operative hypotension and how this was largely inadequately treated. Inmajor surgical cases or cases in which there is expected to be large fluidlosses, invasive monitoring of blood pressure and central venous pressureshould be instituted. There should be earlier use of inotropic cardio-vascular support when hypotension fails to respond to fluid loading.

The choice of anaesthetic technique depends on the type of surgery proposed,the physical status of the patient and patient preference. A recent meta-analysissuggests that regional techniques alone or combined with GA significantlyreduces post-operative morbidity.23 A regional technique should be considered forlimb, perineum and lower abdominal surgery and for laparotomy when combinedwith GA.

THE ELDERLY PATIENT

111

Chap-07.qxd 2/2/02 12:59 PM Page 111

When choosing GA in the elderly, the following should be considered:

• Edentulous patients may present a difficult airway once anaesthesia isinduced as the face ‘collapses’ making a seal with the facemask, andtherefore ventilation difficult. Cervical spondylosis may make intub-ation difficult as neck extension is reduced.

• All elderly patients should be preoxygenated prior to induction ofanaesthesia. Intravenous induction agents should be given slowly.In general the induction dose is lower and induction time prolonged.The MAC of inhalational agents is reduced but the dose of both depolarising and non-depolarising muscle relaxants is the same as ayounger adult.

• The elderly are more sensitive to opioid analgesics but have delayedelimination and so doses should be reduced and dosing interval prolonged.

• Inhalational anaesthetic agents all depress the ventilatory responses tohypoxia and hypercarbia and this will be exacerbated in the elderly whoalready have blunted responses to changes in oxygen and carbon dioxidelevels. All elderly patients should receive supplementary oxygen in therecovery room and probably continued on the ward.

Regional anaesthesia (spinal subarachnoid block and epidural) can be used aloneor in conjunction with GA. It is the author’s belief that spinal blockade should be used alone and that only epidural blockade is combined with GA. Points toconsider in the elderly for regional anaesthesia are:

• Informed consent must be obtained from the patient. The only excep-tion to this is where the patients cannot give consent (e.g. seniledementia) and it is felt that a regional technique is in the best interestsof the patient (e.g. fractured neck of femur).

• Conditions that lead to a fixed cardiac output (e.g. aortic stenosis) andsignificant coagulopathy are excluded. A number of elderly patients areon low dose aspirin (� 300 mg) and this is generally deemed not to bea contraindication to regional anaesthesia.24

• Regional anaesthesia may be more technically difficult in the elderlydue to osteoarthritis, kyphoscoliosis and osteoporotic collapse. Vertebralcollapse means that the spinal cord ends at a lower vertebral level in the elderly and is at risk of damage if the L3/4 space is used. A recentstudy has shown that there is a great variability between the surfacelocalisation of the L3/4 space and the true space25 and a case report hashighlighted the risk to the spinal cord when the wrong interspace isidentified.26


112

Chap-07.qxd 2/2/02 12:59 PM Page 112

• Sympathetic blockade reduces cardiac preload and in the elderly mayresult in profound hypotension which must be treated promptly andaggressively with fluids and vasoconstrictors.

Post-operative care

Post-operative care for any patient involves four basic principles, namely the post-operative visit, post-operative analgesic regimen, fluid and oxygen therapy andpost-operative placement of the patient.

• The post-operative visit is important to the anaesthetist and the patient.It should be performed on everyone and be viewed as the opportunityto review the patient and check that post-operative instructions havebeen followed. If the patient is not progressing as well as expected itaffords time to institute more aggressive management and perhapstransfer to a higher dependency level.

• Fluid prescription post-operatively will depend upon the nature of theprocedure performed, the expected ongoing losses and the expectedperiod that oral intake will be limited. Any prescription must take intoaccount the volume of ongoing loss as well as the daily maintenancerequirements. A well organised fluid balance chart is invaluable.Ongoing losses that are extracellular should be replaced with a balancedsalt solution such as compound sodium lactate. Maintenance fluids canbe roughly calculated from 60 ml/hr for the first 30 kg body weightplus 1 ml/hr for each kg thereafter and should total 1 mmol/kg of Na� and K� every 24 h.

• Oxygen prescription also depends on the nature of the procedure andthe pre-existing medical condition of the patient. Supplemental oxygenshould be prescribed for those who have had thoracic or abdominalsurgery, a history of ischaemic heart disease or respiratory insufficiency.The duration of oxygen therapy is determined on an individual basis sothat a patient with angina having had gastric surgery should receiveoxygen for at least 72 h after surgery. Any patient with a patient con-trolled analgesia (PCA) device should receive oxygen for the durationof use of the PCA.

• Analgesic regimens will be tailored to the type of surgery and physicalstatus of the patient. Non-steroidal anti-inflammatory agents should beused with care in those with borderline renal function. If opioids areused then the dosing interval should be increased. Elderly patients canbe safely given a PCA device on the ward, but should only receive oneif they have the understanding and dexterity to use it. If the hospital hasan acute pain team then patients with epidurals may be safely nursed on

THE ELDERLY PATIENT

113

Chap-07.qxd 2/2/02 12:59 PM Page 113

the general surgical ward. If no facility exists they should be cared foron a HDU as the elderly are more likely to get an inadvertent highblock than younger patients.

• Age should not be a discriminator to admission to a HDU or ICU.Indeed if it is felt that major surgery will be of benefit to the patientthen it seems perverse to deny them appropriate post-operative care.A recent debate in the literature was provoked by a case report27 thatdocumented the pre- and post-operative care of a 113 year old on anICU. The majority of aged patients will be adequately cared for on ageneral surgical ward but a few will require post-operative HDU orICU care which, providing that surgery was appropriate, should bereadily available.

Further reading

Silverstein J. Geriatrics. Anesthesiol Clin North Am 2000; 18: 1–209.

Jin F, Chung F. Minimizing perioperative adverse events in the elderly. Br JAnaesth 2001; 87: 608–24.

McConachie I. Anaesthesia for the senior citizen. Hospital Update 1996; 22: 82–91.

References

1. The Office for National Statistics Population Trends. Winter 2000, 2000. London:The Stationary Office.

2. Rodeheffer RJ, Gerstenblith G, Becker LC et al. Exercise cardiac output ismaintained with advancing age in healthy human subjects: cardiac dilatationand increased stroke volume compensate for a diminished heart rate.Circulation 1984; 69 (2): 203–13.

3. Lakatta EG. Changes in cardiovascular function with aging. Eur Heart J 1990;11 (suppl. C): 22–9.

4. Folkow B, Svanborg A. Physiology of cardiovascular aging. Physiol Rev 1993;73: 725–64.

5. Lakatta EG. Cardiovascular regulatory mechanisms in advanced age. PhysiolRev 1993; 73: 413–67.

6. Priebe H-J. The aged cardiovascular risk patient. Br J Anaesth 2000; 85 (5):763–78.

7. Peterson DD, Pack AI, Silage DA et al. Effects of aging on ventilatory andocclusion pressure responses to hypoxia and hypercapnia. Am Rev Respir Dis1981; 124: 387–91.


114

Chap-07.qxd 2/2/02 12:59 PM Page 114

8. Wahba WM. Influence of aging on lung function – clinical significance ofchanges from age twenty. Anesth Analg 1983; 62: 764–76.

9. Crapo RO,Campbell EJ. Aging of the respiratory system. In Fishman AP (ed.),Pulmonary Diseases and Disorders. New York, NY: McGraw-Hill, 1998; 251–64.

10. McLachlan MSF. The ageing kidney. Lancet 1978; 2: 143–6.

11. Lindeman RD. Renal physiology and pathophysiology of aging. Contrib Nephrol1993; 105: 1–12.

12. Kampmann JP, Sinding J, Moller-Jorgensen I. Effect of age on liver function.Geriatrics 1975; 30: 91–95.

13. Woodhouse KW, Mutch E, Williams FM et al. The effect of age on pathwaysof drug metabolism in human liver. Age Ageing 1984; 13: 328–34.

14. Creasy H, Rapoport SI. The ageing human brain. Ann Neurol 1985; 17: 2–10.

15. Dorfman LJ, Bosley TM. Age-related changes in peripheral and central nerveconduction in man. Neurology 1979; 29: 38–44.

16. Collins KJ, Exton-Smith AN, James MH. Functional changes in autonomicnervous responses with ageing. Age Ageing 1980; 9: 17–24.

17. Montamat SC, Cusack BJ, Vestal RE. Management of drug therapy in the elderly. N Engl J Med 1989; 231 (5): 303–9.

18. Variability in drug response. In Calvey TN, Williams NE (eds), Principles andPractice of Pharmacology for Anaesthetists. Oxford: Blackwell ScientificPublications, 1991; 133–5.

19. Extremes of age. The 1999 report of the National Confidential Enquiry intoPerioperative Deaths. National CEPOD ISBN 0 95222069 6 X.

20. Then and now. The 2000 report of the National Confidential Enquiry intoPerioperative Deaths. National CEPOD ISBN 0 9522069 7 8.

21. Kurz A, Plattner O, Sessler DI et al. The threshold for themoregulatory vaso-constriction during nitrous oxide/isoflurane anesthesia is lower in elderly thanin young patients. Anesthesiology 1993; 79: 465–9.

22. Frank SM, Shir Y, Raja SN et al. Core hypothermia and skin–surface temper-ature gradients. Epidural versus general anesthesia and the effects of age.Anesthesiology 1994; 80: 502–8.

23. Rodgers A,Walker N, Schug S et al. Reduction in postoperative mortality andmorbidity with epidural or spinal anaesthesia: results from overview of ran-domised trials. Br Med J 2000; 321: 1493.

24. Knowles PR. Central nerve block and drugs affecting haemostasis – are theycompatible? Curr Anaesth Crit Care 1996; 7: 281–8.

THE ELDERLY PATIENT

115

Chap-07.qxd 2/2/02 12:59 PM Page 115

25. Broadbent CR, Maxwell WB, Ferrie R et al. Ability of anaesthetists to iden-tify a marked lumbar interspace. Anaesthesia 2000; 55 (11): 1122–6.

26. Greaves JD. Serious spinal cord injury due to haematomyelia caused by spinalanaesthesia in a patient treated with low-molecular weight heparin.Anaesthesia 1997; 52: 150–4.

27. Oliver CD, White SA, Platt MW. Surgery for fractured femur and electiveICU admission at 113 yr of age. Br J Anaesth 2000; 84 (2): 260–2.


116

Chap-07.qxd 2/2/02 12:59 PM Page 116

117

8PERIOPERATIVE OPTIMISATION

Death rates from surgery and anaesthesia are now low.This is despite increasingcomplexity of the surgery performed and a rising average age of the population.However, just a small percentage of the patients undergoing surgery still carrymost of the postoperative morbidity and mortality:

• Those most at risk are the elderly and those undergoing emergencysurgery.

• Coexisting disease such as cardio-respiratory disease or diabetes mellitusfurther increases the chance of death.

The cause of death is most commonly from myocardial infarction or the gradualdevelopment of multiple organ failure syndrome (MODS), the median day ofdeath being on the sixth postoperative day.The mechanisms of the developmentof MODS are still being elucidated.However, it is likely that it results from inflam-matory cascades provoked by a multi-factorial aetiology that may include anycombination of

• altered microcirculation causing tissue injury;

• ischaemia–reperfusion injury;

• direct surgical or traumatic tissue injury;

• surgical stimulation of metabolic and endocrine processes;

• blood loss and fluid shifts causing regional and global hypoperfusion;

• anaesthetic agents causing vasodilatation and altered regional blood flow;

• splanchnic hypoperfusion – this may be especially important sincesplanchnic blood may amount to two-fifths of the blood volume; dam-age to mucosal integrity and bacterial translocation may be importantin initiating inflammatory cascades.

Chap-08.qxd 2/1/02 12:07 PM Page 117

Many of these perioperative events have potential to cause an imbalance betweenoxygen delivery and demand, be it local or global.This is especially likely to occurin the presence of reduced physiological reserve where cardiac index (CI) cannotrise to meet the demand placed by surgery. The tissue hypoxia that results may precipitate the systemic inflammatory response syndrome (SIRS) that may thenprogress on to MODS.

If this is the case, then to reduce perioperative risk,we need to target those patientswith limited physiological reserve and undergoing surgery of sufficient physio-logical insult.

IDENTIFYING PERIOPERATIVE RISK

In the scoring systems of Goldman and Detsky (see Chapter 1), particular risk isattached to poor cardiac reserve in the form of congestive cardiac failure, aorticstenosis and precarious myocardial perfusion.

This suggests the ability to meet the demands of surgery by maintaining or increas-ing perfusion and oxygen delivery might be important in determining survival:

• In support of this, early work by Boyd et al.1 had suggested in cardiacsurgery that failure to raise postoperative CI above 2.5 l/min/m2 wasassociated with increased mortality rate.

• Similarly,Clowes et al.2 later showed that failure to increase cardiac out-put after thoracic surgery was associated with reduced survival.

• Shoemaker3 also defined indicators of risk for perioperative death bycorrelating preoperative criteria with mortality rates (table 8.1). Thiswork identified both patient criteria and criteria relating to the type of surgery to be undertaken.They and others subsequently used thesecriteria to identify and study high-risk patients.


118

Table 8.1 – Shoemaker’s indicators of high risk.

Previous severe cardio-respiratory illness, e.g. acute myocardial infarction or chronic obstructive airways disease.

Extensive ablative surgery planned for malignancy, e.g. gastrectomy, oesophagectomy or surgery � 6 h.Multiple trauma, e.g. more than three organ injuries, more than two systems or opening two body cavities.Massive acute haemorrhage, e.g. � 8 units.Age above 70 years and limited physiological reserve of one or more organs.Septicaemia (positive blood cultures or septic focus) WCC � 13, pyrexia to 38.3 for 48 h.Respiratory failure (pO2 � 8 kPa on an FiO2 � 0.4 or mechanical ventilation � 48 h.Acute abdominal catastrophe with haemodynamic instability (e.g. pancreatitis, perforated viscus, peritontis, gastrointestinal bleed).

Acute renal failure (urea � 20 mmol/l, creatinine � 260 mmol/l).Late stage vascular disease involving aortic disease.Shock, e.g. MAP � 60 mmHg, CVP � 15 cmH2O, urine output � 20 ml/h.

Chap-08.qxd 2/1/02 12:07 PM Page 118

Improving outcome

Having identified the high-risk patient, how do we then set about improving outcome?

There are several potential strategies to consider:

• Treatment of existing medical disease – the principle of treating treatablemedical conditions such as congestive cardiac failure, hypertension andrespiratory disease is well established in anaesthetic practice, althoughthe evidence for benefit, e.g. in moderate degrees of hypertension islimited.

• Resuscitation of presenting disease – where time allows attention should be directed to correcting electrolyte, metabolic and fluid balance.Circulating volume status and blood pressure should be restored usingtitrated fluid and inotropes.

• Use of regional anaesthesia – the use of regional anaesthesia has beenshown to result in improved postoperative respiratory function. To date, however, there is little evidence to suggest that mortality rates areaffected.

• Strategies to prevent myocardial events – beta blockers, nitrates, calciumchannel blockers and alpha2 antagonists have all been used to try toreduce postoperative mortality. In patients with ischaemic heart disease,there is some evidence of reductions in perioperative ischaemic eventsfollowing administration of beta blockers.4

• Cardio-respiratory optimisation – there is increasing evidence to supportthe use of invasive monitoring, titrated fluids and inotropes to achieveenhanced cardiovascular function in anticipation of increased perioper-ative oxygen demand.

This evidence will be reviewed in the remainder of this chapter.

THE EVIDENCE FOR PERIOPERATIVE OPTIMISATION

The perioperative cardiovascular changes seen in high-risk patients undergoingmajor surgery were characterised in the 1970s by Shoemaker:

• In 1973, his group studied 98 patients undergoing major surgery and in variable levels of established shock. Comparing 67 survivors with 31 non-survivors, they were able to demonstrate significantly differenthaemodynamic changes in the days following surgery. Surviving patientshad a higher CI, higher oxygen delivery and higher oxygen uptake.These indices were much better predictors of mortality than the moretraditionally used values of blood pressure and heart rate.5

PERIOPERATIVE OPTIMISATION

119

Chap-08.qxd 2/1/02 12:07 PM Page 119

• Following this, the same group studied 220 high-risk patients this timeundergoing major-elective and semi-elective surgery. In these patients,they confirmed higher values of CI, oxygen delivery and oxygen con-sumption in the survivors. From these findings, they postulated that ifcardiovascular performance could be enhanced in high-risk patients toachieve the CI and oxygen delivery values manifest by survivors thenoverall survival rate could increase.They were able to suggest specificgoals of CI (4.5 l/min/m2),oxygen delivery (600ml/min/m2) and oxygenconsumption (170 ml/min/m2) that they termed ‘supranormal values’.6

• In a subsequent non-randomised study of 100 patients, they eitheractively increased cardiovascular performance with fluids and inotropesaiming to achieve the above supranormal values or in the control patientsallowed CI to remain between 2.8 and 3.5 l/min/m2. Mortality andcomplication rate were both reduced in the intervention group.7

Several other non-randomised studies confirmed these findings, however, it wasnot until the 1980s that further evidence was gained from properly randomisedcontrolled trials:

• Firstly, in 1985 Schultz et al.8 studied 70 patients undergoing surgicalrepair of hip fractures. Half of the patients were monitored with pul-monary artery flotation catheters and managed with intravenous fluidsand inotropes to enhance CI and oxygen delivery to preset goals.Theother half were managed conventionally. The intervention group had asignificantly lower mortality rate by over 25%. It was unclear, however,whether the improvement was due to better monitoring or to enhancedoxygen delivery.

• Subsequently, in 1988 Shoemaker’s group9 published a randomisedcontrolled trial of 340 high-risk surgical patients. They recruitedpatients using their previously defined criteria for high risk. Controlpatients were managed conventionally, whereas the protocol groupwere given intravenous fluids, inotropes and vasodilators aiming toachieve the supranormal values they had described previously. The protocol group had significantly lower mortality (4% vs 33% p � 0.01)and complication rate (1.3 vs 0.4 p � 0.05).

• Another group, Berlauk et al.,10 published further data to support theuse of perioperative optimisation. In this study, 89 patients underwentperipheral vascular surgery under general anaesthesia. Patients wererandomised into three groups. One group received conventional man-agement and the other two groups were monitored with a pulmonaryartery catheter either placed 12 h before or immediately before surgery.In the latter two groups, the invasive monitoring was used to guide


120

Chap-08.qxd 2/1/02 12:07 PM Page 120

circulatory ‘tune up’ with fluid loading, inotropes and vasodilators.Treatment was given to maintain pulmonary capillary wedge pressureof 8–15 mmHg, CI � 2.8 l/min/m2 and SVR � 1100 dyne s/cm5 pre-and intraoperatively. The study groups both had fewer intraoperativeadverse events, less early graft thrombosis and lower postoperative cardiac morbidity.

• In 1992 and 1995,Shoemaker’s group published two further randomisedcontrolled trials this time in 67 and 125 patients with severe trauma.11,12

In patients treated to increase oxygen delivery to supranormal values,they found reduced organ failure, lower mortality rate, shorter stays inintensive care and shorter periods of ventilation.

• In a further randomised controlled trial, Boyd et al.13 studied 107patients at high risk as defined by Shoemaker’s criteria.The majority ofpatients were admitted to intensive care preoperatively, although somewere admitted postoperatively.All patients received conventional therapywith invasive monitoring, intravenous fluids, vasodilators and inotropes.The protocol group were managed in addition to deliberately increasedoxygen delivery. Dopexamine was used to achieve goals of oxygendelivery of � 600 ml/min/m2 with pulmonary capillary wedge pressure12–14 mmHg and haemoglobin � 12 g/l.The result was a significantlylower mortality and complication rate (by 75% and 59%, respectively).

These impressive reductions in mortality and complication rates have not, how-ever, been mirrored in some more recent randomised trials. Ziegler et al.14 andValentine et al.15 both failed to show any significant benefit in terms of morbidityand mortality from optimisation attempts in patients undergoing peripheral vascular surgery.

Another recent study16 of over 400 patients undergoing abdominal surgery usingdopexamine also failed to show benefit.

It is possible that in at least some of these studies the targeted populations were notat high enough risk and therefore unlikely to show benefit from optimisationstrategies.Clearly, the patients chosen for optimisation protocols need to be selectedwith care.Further, in one of these studies15 the intraoperative complication rate wasactually increased in the optimised group. Optimisation techniques should there-fore be titrated with care to avoid inducing morbidity related to the therapy.

Another recent study by Wilson et al.17 has added an exciting dimension to theconcept of perioperative optimisation. In a randomised controlled trial of 138patients they studied three high-risk groups undergoing surgery:

• The control group were managed conventionally and were admitted tointensive care if deemed necessary.


121

Chap-08.qxd 2/1/02 12:07 PM Page 121

• The other two groups were admitted preoperatively to intensive careand given goal directed therapy with either adrenaline or dopexaminelasting for at least 12 h postoperatively.

• Both the treatment groups had significantly improved survival rate.

• However, only the dopexamine group saw a significant reduction inmorbidity.

This is particularly interesting because the dopexamine treated group did not seean increase in CI by as much as the adrenaline treated group.The reduced mor-bidity was due to a reduction in sepsis and ARDS.

Dopexamine is a pure beta2 agonist with very little beta1 effect and no alpha1

activity. Its inotropic effects result from inhibition of endogenous catecholaminereuptake and from stimulation of baroreceptor reflexes. It reduces systemic andpulmonary vascular resistance and increases both renal and splanchnic blood flow.Beta2 stimulation also brings about anti-inflammatory properties. It may be thatthis together with improved splanchnic blood flow both maintains mucosal bar-riers and attenuates the inflammatory cascade that leads to SIRS and MODS.In support of this, dopexamine has been shown to reduce inflammatory change inthe upper gastrointestinal mucosa in high-risk surgical patients.18

The specific benefits of dopexamine suggested in the study of Wilson et al. are byno means universally accepted. It has been suggested19 that the three treatmentgroups in the study were not comparable and that this may have led to some of thedifferences seen between the dopexamine and adrenaline treated groups. Furtherwork is therefore needed to establish the place of dopexamine.

The role of catecholamines and their actions at different adrenoceptors on theimmune system is explored fully in a review cited in Further reading.

PATIENT SELECTION

There is accumulating evidence to suggest that the use of perioperative optimisa-tion can improve postoperative outcome.The challenge is to identify

• those patients who will benefit,

• the appropriate goals to direct therapy to.

Table 8.1 lists Shoemaker’s criteria for high-risk patients. However, we have seen that where perioperative risk is low there is little benefit from optimisationregimens. In fact, some studies have seen increased risk in optimised patients.In selecting patients who may benefit from optimisation, the anaesthetist must balance the risks of surgery and anaesthesia against possible detrimental effects ofoptimisation.


122

Chap-08.qxd 2/1/02 12:07 PM Page 122

Shoemaker has also suggested therapeutic goals to target therapy to, the supranormalvalues described previously. Others have suggested that patients with initial oxygendelivery of � 450 ml/min/m2 should be targeted.Again it is probably a matter ofjudgement on the part of the anaesthetist to enhance cardiovascular function sufficiently to allow the demands of surgery to be met whilst at the same time notoverstressing the myocardium and precipitating ischaemia or arrhythmia.

OPTIMISATION IN CRITICAL ILLNESS

Although there is good evidence to support the use of preoperative optimisationin elective and semi-elective surgery, there is unfortunately little to support the pursuit of supranormal goals for oxygen delivery in patients who are alreadycritically ill:

• Initial non-randomised studies in patients with established septicshock20 suggested that optimisation might be of benefit, however,subsequent randomised controlled trials have not supported this.

• A randomised study of patients with septic shock by Tuchshmidt et al.21

found a reduced mortality rate in optimised patients but this was notsignificantly so.

• Other randomised studies22–24 have all failed to show benefit fromoptimisation. All of these studies had in common the recruitment ofpatients with established septic shock and/or organ failure.

Another interesting study helps to clarify the situation:

• Gutierrez et al.25 examined pHi as indicator of hypoperfusion at entry totheir study.Those with a low pHi at entry to the study, suggesting exist-ing hypoperfusion, did not benefit from optimisation whereas thosewithout hypoperfusion at entry to the study did see reduced mortality.

It would appear that where organ failure is already established, there is little to gainfrom pursuing supranormal values.

Further reading

Uusaro A, Russell JA. Could anti-inflammatory actions of catecholamines explainthe possible beneficial effects of supranormal oxygen delivery in critically illsurgical patients? Inten Care Med 2000; 26: 299–304.

Kelly KM. Does increasing oxygen delivery improve outcome? Yes. Crit Care Clin1996; 12: 635–44.

Ronco JJ, Fenwick JC,Tweeddale MG. Does increasing oxygen delivery improveoutcome in the critically ill? No. Crit Care Clin 1996; 12: 645–59.


123

Chap-08.qxd 2/1/02 12:07 PM Page 123

References

1. Boyd AR, Tremblay RE, Spencer FC et al. Estimation of cardiac output soon after intracardiac surgery with cardio-pulmonary bypass. Ann Surg 1959;150: 613.

2. Clowes GHA, Del Guercio LRM. Circulatory response to trauma of surgicaloperations. Metabolism 1960; 9: 67.

3. Shoemaker WC, Czer LS. Evaluation of the biologic importance of varioushemodynamic and oxygen transport variables: which variables should bemonitored in postoperative shock? Crit Care Med 1979; 7: 424.

4. Poldermans D, Boersma E, Bax JJ et al. The effect of bisoprolol on peri-operative mortality and myocardial infarction in high-risk patients under-going vascular surgery. Dutch Echocardiographic Cardiac Risk EvaluationApplying Stress Echocardiography Study Group. N Engl J Med 1999; 341:1789–94.

5. Shoemaker WC, Montgomery ES, Kaplan E et al. Physiologic patterns in sur-viving and nonsurviving shock patients. Use of sequential cardiorespiratoryvariables in defining criteria for therapeutic goals and early warning of death.Arch Surg 1973; 106: 630.

6. Shoemaker WC, Pierchala C, Chang P et al. Prediction of outcome and sever-ity of illness by analysis of the frequency distributions of cardiorespiratoryvariables. Crit Care Med 1977; 5: 82.

7. Shoemaker WC,Appel PL,Waxman K et al. Clinical trial of survivors’ cardio-respiratory patterns as therapeutic goals in critically ill postoperative patients.Crit Care Med 1982; 10: 398.

8. Schultz RJ,Whitfield GF, LaMura et al.The role of physiologic monitoring inpatients with fractures of the hip. J Trauma 1985; 25: 309.

9. Shoemaker WC, Appel PL, Kram HB et al. Prospective trial of supranormalvalues of survivors as therapeutic goals in high-risk surgical patients. Chest1988; 94: 1176.

10. Berlauk JF,Abrams JH, Gilmour IJ et al. Preoperative optimization of cardio-vascular hemodynamics improves outcome in peripheral vascular surgery.A prospective, randomized clinical trial. Ann Surg 1991; 214: 289.

11. Fleming A, Bishop M, Shoemaker W et al. Prospective trial of supranormal values as goals of resuscitation in severe trauma. Arch Surg 1992; 127: 1175.

12. Bishop MH, Shoemaker WC,Appel PL et al. Prospective, randomized trial ofsurvivor values of cardiac index, oxygen delivery, and oxygen consumption asresuscitation endpoints in severe trauma. J Trauma 1995; 38: 780.


124

Chap-08.qxd 2/1/02 12:07 PM Page 124

13. Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risksurgical patients. JAMA 1993; 270: 2699.

14. Ziegler DW,Wright JG, Choban PS et al. A prospective randomized trial ofpreoperative ‘optimization’ of cardiac function in patients undergoing electiveperipheral vascular surgery. Surgery 1997; 122: 584.

15. Valentine RJ, Duke ML, Inman MH et al. Effectiveness of pulmonary arterycatheters in aortic surgery: a randomized trial. J Vasc Surg 1998; 27: 203.

16. Takala J, Meier-Hellmann A, Eddleston J et al. Effect of dopexamine on out-come after major abdominal surgery: a prospective, randomized, controlledmulticenter study. European Multicenter Study Group on Dopexamine inMajor Abdominal Surgery. Crit Care Med 2000; 28: 3417.

17. Wilson J,Woods I, Fawcett J et al. Reducing the risk of major elective surgery:randomised controlled trial of preoperative optimisation of oxygen delivery.Br Med J 1999; 318: 1099.

18. Byers RJ, Eddleston JM, Pearson RC et al. Dopexamine reduces the incidenceof acute inflammation in the gut mucosa after abdominal surgery in high-riskpatients. Crit Care Med 1999; 27: 1787.

19. Snowden C,Roberts D.More than an abstract:‘To optimise or not to optimisethat is the question’. CPD Anaes 2000; 2: 43.

20. Edwards JD,Brown GC,Nightingale P et al.Use of survivors’ cardiorespiratoryvalues as therapeutic goals in septic shock. Crit Care Med 1989; 17: 1098.

21. Tuchschmidt J, Fried J,Astiz M et al. Elevation of cardiac output and oxygendelivery improves outcome in septic shock. Chest 1992; 102: 216.

22. Hayes MA,Yau EH, Timmins AC et al. Response of critically ill patients totreatment aimed at achieving supranormal oxygen delivery and consumption.Relationship to outcome. Chest 1993; 103: 886.

23. Yu M, Levy MM, Smith P et al. Effect of maximizing oxygen delivery on morbidity and mortality rates in critically ill patients: a prospective, randomized,controlled study. Crit Care Med 1993; 21: 830.

24. Gattinoni L, Brazzi L, Pelosi P et al. A trial of goal-oriented hemodynamictherapy in critically ill patients. SvO2 Collaborative Group.N Engl J Med 1995;333: 1025.

25. Gutierrez G, Palizas F, Doglio G et al. Gastric intramucosal pH as a therapeuticindex of tissue oxygenation in critically ill patients. Lancet 1992; 339: 195.


125

Chap-08.qxd 2/1/02 12:07 PM Page 125

Chap-08.qxd 2/1/02 12:07 PM Page 126


127

9THE PATIENT WITH CORONARY

HEART DISEASE

Coronary heart disease (CHD) is widely prevalent in the adult population of theUnited Kingdom and is a major cause of morbidity and mortality:1

• In the surgical population, cardiovascular disease is the leading cause ofdeath within 30 days of a surgical procedure, responsible for 36% ofperi-operative deaths in the latest NCEPOD report.2

• Longer-term morbidity and mortality is also increased in patients withor at risk for CHD who undergo non-cardiac surgery.3

The management of the patient with CHD who requires anaesthesia for non-cardiac surgery is therefore an important issue for all anaesthetists.

PATHOPHYSIOLOGY

The major pathological feature in the patient with CHD is the presence of lipidatheromatous plaques within the walls of the epicardial coronary arteries. This feature, exacerbated sometimes by coronary arterial spasm, may lead to:

• Myocardial ischaemia when myocardial oxygen demand exceeds a limitedmyocardial oxygen supply; manifest clinically as stable angina.

• Rupture of plaque with thrombus formation leading to near or com-plete coronary arterial occlusion; manifest clinically as unstable angina,myocardial infarction or sudden cardiac death (the acute coronary syndromes).

In the patient who undergoes surgery, a number of profound stresses, includingsurgical trauma to tissue,marked changes in temperature, alteration of major organfunction and the stress and inflammatory responses, combine to produce a rangeof pathophysiological derangements which, in the presence of CHD, may lead tomyocardial injury in the peri-operative period.

Chap-09.qxd 2/2/02 1:01 PM Page 127

• Sympathetic activity may result in myocardial oxygen supply/demandimbalance leading to myocardial ischaemia, and/or plaque instabilityand rupture.

• The inf lammatory response may result in activation of vascularendothelium leading to plaque instability and rupture, and/or coronaryarterial spasm.

• Changes in platelet function, and in the coagulation and fibrinolyticsystems, predispose to coronary arterial thrombosis.

The patient with CHD who requires major surgery is thus at risk of myocardialinjury which may become manifest in the immediate or delayed post-operativeperiod as myocardial ischaemia or infarction, serious arrhythmias, ventricular failureor sudden cardiac death.

PRE-OPERATIVE MANAGEMENT

Aims

• Identification of the patient with severe CHD.

• Consideration of revascularisation and/or modification of medical therapy.

Identification of the patient with severe CHD

Guidelines for the pre-operative cardiac assessment of patients undergoing non-cardiac surgery have been published by the American College of Cardiology inconjunction with the American Heart Association.4 A cardiological perspective onthese matters is given in a later chapter:

• The identification of the patient with severe CHD is based upon anevaluation of information obtained from the history, examination andresting electrocardiogram.

• The identification of a patient with an acute coronary syndrome clearlymandates the postponement of elective or scheduled surgery and theimmediate introduction of the appropriate medical therapy.

• In the majority of patients, however, the pre-operative assessmentallows the identification of clinical markers associated with the presenceof underlying severe CHD and with a high peri-operative and long-term cardiac risk.

The following clinical markers are most important:5

• symptoms of myocardial ischaemia (stable angina),

• history of previous myocardial infarction,


128

Chap-09.qxd 2/2/02 1:01 PM Page 128

• congestive cardiac failure,

• diabetes mellitus.

In patients with several or all of these clinical markers, there is a high probabilityof severe CHD (left main stem disease, three vessel disease, or two vessel diseasewith involvement of the proximal left anterior descending artery), whilst inpatients with none of these clinical markers, the risk of severe underlying CHD isless than 5%.6 The effect of these clinical markers on peri-operative and long-termcardiac risk is similar – in patients undergoing vascular surgery who have none ofthese clinical markers, the combined incidence of peri-operative myocardialinfarction and cardiac death is approximately 3%, whilst in patients with three ormore of these markers, the risk is between 15% and 20%.7

The requirement for further cardiac investigation in a patient with the above clin-ical markers for CHD should be determined by a balanced assessment of the func-tional capacity of the patient and the cardiac risk (risk of myocardial infarction orcardiac death) associated with the specific surgical procedure to be performed.

The cardiac risk for non-cardiac surgical procedures has been stratified into high,intermediate and low risk groups:4

High cardiac risk (� 5% mortality):

• major emergency surgery, particularly in the elderly,

• aortic and peripheral vascular surgery,

• anticipated prolonged procedures associated with large fluid shiftsand/or blood loss.

Intermediate cardiac risk (� 5% mortality):

• carotid endarterectomy,

• head and neck surgery,

• intraperitoneal and intrathoracic surgery,

• orthopaedic surgery,

• prostate surgery.

Low cardiac risk (� 1% mortality):

• endoscopic procedures,

• superficial procedures,

• cataract surgery,

• breast surgery.

THE PATIENT WITH CORONARY HEART DISEASE

129

Chap-09.qxd 2/2/02 1:01 PM Page 129

In general, in the presence of clinical markers for severe CHD, a patient who is toundergo high risk surgery requires further pre-operative cardiac investigation,whilst a patient who is to undergo low risk surgery does not require further pre-operative cardiac investigation.

Further pre-operative cardiac investigation is based on the identification of reversiblemyocardial ischaemia by one of a variety of methods of non-invasive testing:

• exercise stress electrocardiography,

• pharmacological stress electrocardiography,

• ambulatory electrocardiography,

• stress echocardiography,

• myocardial perfusion imaging.

In most ambulatory patients, the investigation of choice is exercise stress electro-cardiography, which can provide an estimate of functional capacity and detectmyocardial ischaemia through changes in the electrocardiograph and the haemo-dynamic response. Pharmacological stress electrocardiography is appropriate inthose patients unable to exercise for non-cardiac reasons, and stress echocardio-graphy or myocardial perfusion imaging appropriate in those patients with anabnormality of the resting electrocardiogram such as bundle branch block.

Invasive testing by coronary angiography is appropriate in the following groups ofpatients who are to undergo non-cardiac surgery:8

• patients with a high risk result during non-invasive testing,

• patients with myocardial ischaemia unresponsive to adequate medicaltherapy,

• patients with an acute coronary syndrome,

• patients with equivocal non-invasive testing who are to undergo highrisk non-cardiac surgery.

The adoption of this structured approach will identify the majority of patientsrequiring non-cardiac surgery who have severe CHD. However, there remains noreasonable way to absolutely eliminate the possibility that an individual patientmay suffer a cardiac complication during or after a surgical procedure.

Consideration of revascularisation and/or modification of medical therapy

Prophylactic coronary artery bypass grafting (CABG) prior to a non-cardiac elective surgical procedure is appropriate only in those patients who fulfil standard


130

Chap-09.qxd 2/2/02 1:01 PM Page 130

criteria for CABG surgery for prognostic reasons independent of the non-cardiacprocedure:9

• suitable viable myocardium with left main stem disease,

• three vessel coronary artery disease with left ventricular dysfunction,

• two vessel coronary artery disease including left anterior descendingdisease with left ventricular dysfunction,

• myocardial ischaemia unresponsive to maximal medical therapy.

Although successful myocardial revascularisation by CABG surgery in othergroups almost normalises non-cardiac peri-operative risk, this benefit is lost by thecumulative morbidity and mortality of the cardiac and the non-cardiac surgicalprocedures. The role of prophylactic percutaneous transluminal coronary angio-plasty (PTCA) prior to a non-cardiac elective surgical procedure is less clearlydefined as there is no evidence of prognostic benefit for angioplasty over medicaltherapy.10 PTCA, in this setting, should be restricted to patients with reversiblemyocardial ischaemia in whom a single coronary arterial stenosis subtends a largearea of viable myocardium:

• The majority of patients with CHD presenting for elective non-cardiacsurgery do not have disease severe enough to justify the risks of coron-ary angiography and coronary revascularisation.

• In these patients, the optimisation of pre-operative medical therapy andthe continuation of this medical regimen through the operative andpost-operative periods is most appropriate.

Patients with CHD should be receiving one or more of the following medicationsfor symptom control and/or prophylaxis, unless contraindicated:

• beta blockers,

• calcium channel blockers,

• nitrates,

• potassium channel activators,

• aspirin,

• angiotensin converting enzyme inhibitors or angiotensin receptorantagonists for patients with left ventricular dysfunction.

Beta blockers, calcium channel blockers, nitrates and the potassium channel acti-vator nicorandil all limit the degree of myocardial ischaemia during exercise test-ing and may be expected to be of value peri-operatively.11 Aspirin has a proven rolein the primary and secondary prevention of myocardial infarction, mediated by its


131

Chap-09.qxd 2/2/02 1:01 PM Page 131

antiplatelet actions, and may be expected to be useful in the peri-operative periodwhen platelet reactivity is increased.12 Angiotensin converting enzyme inhibitorsand angiotensin receptor antagonists reduce morbidity and mortality in patientswith CHD and left ventricular dysfunction.13

The strongest evidence supports the peri-operative use of beta blockers, whoseintroduction should be considered in all patients with CHD, or risk factors forCHD, who require surgery:

• Beta blockers limit the chronotropic and inotropic effects of the increasedsympathetic activity present in the peri-operative period, reducingmyocardial oxygen requirements and increasing myocardial oxygen supply by increasing diastolic coronary perfusion time.

• They also limit the shear stress across atheromatous plaques in the coro-nary circulation and so may reduce the incidence of plaque rupture andconsequent coronary arterial thrombosis.

• Beta blockers have been demonstrated to reduce the amount of myo-cardial ischaemia detected by ST segment analysis when administeredperi-operatively.14

• Recent studies have indicated that the prophylactic use of beta blockersin the peri-operative period in patients with CHD, or risk factors forCHD, undergoing non-cardiac surgery reduces cardiac risk in theimmediate and delayed post-operative periods.15,16

The direct evidence to support the introduction of calcium channel blockers,nitrates and the potassium channel activator nicorandil in the peri-operative periodis less certain. Furthermore, these medications may produce vasodilatation andreflex tachycardia which may compromise myocardial perfusion in the patientwith CHD undergoing anaesthesia and surgery. These agents should probably bereserved for patients who have previously required these medications for controlof myocardial ischaemia, or for patients who develop myocardial ischaemia aftersurgery despite the appropriate use of a beta blocker.

The use of aspirin in the patient undergoing major surgery is limited by the per-ception that surgical bleeding is increased. However, aspirin has no effect onplatelet aggregation induced by tissue collagen and should not, in theory, increasethe risk of surgical bleeding.17 The prophylactic use of aspirin should be con-sidered, therefore, in the patient with CHD undergoing major surgery.

Angiotensin converting enzyme inhibitors and angiotensin receptor antagonistsare vasodilators, which may interact with anaesthetic agents and techniques toproduce profound hypotension and compromise myocardial perfusion. Thesemedications are best avoided in the peri-operative period.


132

Chap-09.qxd 2/2/02 1:01 PM Page 132

ANAESTHETIC CONSIDERATIONS IN THE PATIENT WITH SEVERE CHD

Aims

• Optimisation of pre-operative status by appropriate investigation, revas-cularisation and introduction of anti-ischaemic medical therapy.

• Preservation of the balance between myocardial oxygen supply anddemand throughout the peri-operative period.

Optimisation of pre-operative status

The optimisation of pre-operative status and the need to continue the appropriatemedical therapy throughout the peri-operative period has been discussed in theprevious section. However, the degree of urgency of the non-cardiac surgical procedure may limit the amount of time that is available for pre-operative investi-gation and management. (NCEPOD classification of surgical urgency is discussedin Chapter 3):

• There is clearly no opportunity in the patient who requires emergencysurgery for pre-operative investigation or therapy, and attention shouldbe directed towards intensive cardiovascular monitoring and preser-vation of the myocardial oxygen supply and demand balance in theoperative and post-operative periods, and to the introduction of appro-priate anti-ischaemic medication in the post-operative period.

• Similarly, in the patient who requires urgent surgery, there is little timefor the optimisation of pre-operative status.

• In the patient with an acute coronary syndrome who requires urgentnon-cardiac surgery, consideration may be given to the use of an intra-aortic balloon pump device, in those centres which have cardiology orcardiac surgery support, in order to reduce myocardial oxygen require-ments and to improve coronary perfusion and, thus, myocardial oxygensupply.18

• There is sufficient time in the patient requiring scheduled or electivesurgery for the appropriate optimisation of pre-operative status.

• It may become necessary to delay scheduled surgery if the patient isfound to have CHD necessitating revascularisation according to thecriteria discussed previously.

• Occasionally, the use of an intra-aortic balloon pump may be consideredin those patients demonstrated to have severe CHD not amenable to revascularisation.


133

Chap-09.qxd 2/2/02 1:01 PM Page 133

• Current recommendations indicate that a patient who has suffered anuncomplicated myocardial infarction may undergo scheduled non-cardiac surgery four weeks after the myocardial infarction providingrigorous haemodynamic monitoring and control is applied throughoutthe peri-operative period.4,19

However, persisting myocardial ischaemia after infarction, which should be soughtby non-invasive testing, is an indication for invasive investigation which will delaysurgery further if revascularisation should be proved necessary.

• The patient who has suffered an uncomplicated myocardial infarctionwho requires elective non-cardiac surgery should wait for at least 3 months before undergoing surgery in order to minimise the risk ofperi-operative myocardial infarction.20

Preservation of myocardial oxygen supply and demand

The essence of anaesthetic management in the patient with severe CHD under-going non-cardiac surgery is the preservation of the balance between myocardialoxygen supply and demand.

Myocardial oxygen supply is determined by

• coronary artery blood flow,

• arterial oxygen content.

Coronary artery blood flow is dependent on

• coronary perfusion pressure (CPP), determined by the aortic diastolicpressure minus the left ventricular end diastolic pressure (LVEDP);

• coronary vascular resistance, determined by blood viscosity, sympathetictone and, most importantly in the patient with CHD, fixed resistancedue to coronary atheromatous disease;

• duration of diastole determined by the heart rate (shorter diastole withfaster heart rates).

Coronary artery blood flow is directly related to the CPP, and inversely related tothe coronary vascular resistance and the heart rate.

Arterial oxygen content is dependent on

• haemoglobin concentration,

• haemoglobin saturation,

• arterial oxygen tension.


134

Chap-09.qxd 2/2/02 1:01 PM Page 134

Myocardial oxygen demand is determined by

• heart rate,

• left ventricular afterload,

• left ventricular preload,

• myocardial contractility.

In the patient with severe CHD, the myocardial oxygen demand must not beallowed to outstrip the supply.

Increased myocardial oxygen Decreased myocardial oxygen demand supply

Tachycardia TachycardiaIncreased afterload Decreased aortic diastolic pressureIncreased preload (LVEDP) Increased LVEDPIncreased contractility Decreased arterial oxygen content

• Tachycardia and an increased LVEDP have greater potential to inducemyocardial ischaemia as both increase demand and reduce supply, thanhypertension which increases demand, but also supply due to the asso-ciated increase in aortic diastolic pressure resulting in an increased CPP.

• Hypotension, with a decreased aortic diastolic pressure, in associationwith tachycardia and/or increased LVEDP is a particularly hazardouscombination of haemodynamic change that must be avoided in thepatient with severe CHD.

Monitoring

An essential prerequisite for the preservation of the balance between myocardialoxygen supply and demand is the establishment of a level of monitoring appro-priate to the disease severity of the patient and the magnitude of the surgery.The Association of Anaesthetists of Great Britain and Ireland has stated that thefollowing patient monitoring devices are essential to the safe conduct of anaesthe-sia for all patients:21

• electrocardiograph,

• non-invasive arterial pressure monitoring,

• pulse oximetry,

• capnography,

• vapour analysis.


135

Chap-09.qxd 2/2/02 1:01 PM Page 135

This level of monitoring is sufficient in the majority of patients with severe CHDundergoing low risk surgery but must be supplemented by additional monitoringfor patients undergoing intermediate or high risk surgery:

• invasive cardiovascular monitoring,

• temperature,

• urine output.

Invasive cardiovascular monitoring includes arterial and central venous pressuremonitoring. The use of pulmonary artery catheterisation is controversial but islikely to be of benefit in the following groups of patients:22

• patients with a recent myocardial infarction,

• patients with significant CHD undergoing high risk surgical procedures,

• patients with CHD and associated left ventricular dysfunction under-going intermediate risk surgical procedures.

Anaesthetic techniques

There is no evidence to demonstrate a consistent advantage of any general anaes-thetic agent or technique over any other, in relation to the risk of peri-operativecardiac morbidity and mortality. Several large studies of patients undergoing CABGhave demonstrated no difference in outcome with differing anaesthetic agents e.g. volatile agents versus narcotic based techniques.

All anaesthetic agents and techniques are associated with cardiovascular effects,and the care with which the chosen agent and/or technique is managed is moreimportant than the choice of agent or technique itself. Particular attention should,however, be directed to control of the haemodynamic changes associated with thefollowing interventions:

• induction of anaesthesia,

• laryngoscopy and tracheal intubation,

• surgical incision and stimulation,

• emergence from anaesthesia,

• tracheal extubation.

Many pharmacological approaches are available to obtund the haemodynamicresponse to airway manipulation, including the use of opioids or cardiovascularmedications such as beta blockers.23 Alternatively, the haemodynamic response toairway manipulation can be minimised by the use of a laryngeal mask airwayrather than an endotracheal tube, when this is otherwise appropriate.


136

Chap-09.qxd 2/2/02 1:01 PM Page 136

Regional anaesthesia may be used as an adjunct or as an alternative to generalanaesthesia. However, the available evidence in high risk patients undergoinglower extremity vascular surgery indicates that carefully conducted epidural anaes-thesia and general anaesthesia are associated with comparable rates of cardiac morbidity.24 If a regional anaesthetic technique is to be employed, it is essentialthat attention is directed to the avoidance of sudden and severe hypotension andthe potential decrease in coronary perfusion associated with the technique, and consequently, epidural catheter techniques are to be preferred to subarachnoid single-shot techniques.

Detection of intra-operative myocardial ischaemia

Methods of peri-operative surveillance for myocardial ischaemia include:

• computerised ST segment monitoring,

• pulmonary artery pressure monitoring,

• transoesophageal echocardiography.

Computerised ST segment trend analysis is superior to visual interpretation of STsegment changes for the detection of intra-operative and post-operative myocardialischaemia and should be used if available. Similarly, changes in pulmonary arterypressure and pulmonary artery occlusion pressure and waveform can be sensitiveindicators of myocardial ischaemia. Transoesophageal echocardiography, by thedetection of wall motion abnormalities, is a further method in appropriatelytrained hands for the detection of intra-operative myocardial ischaemia.

Management of intraoperative ischaemia

Management of intra-operative myocardial ischaemia should concentrate on:

• The correction of haemodynamic status by manipulation of the depthof anaesthesia and the use of vasoactive agents. Beta blockers may beused to slow the heart rate and, thereby, to limit myocardial oxygendemand and increase myocardial oxygen supply. The rapid onset andtitratability of the intravenous beta blocker esmolol make this a particu-larly suitable agent for use in this role.

• The use of intravenous nitroglycerin which appears to redistributemyocardial blood flow and may reverse myocardial ischaemia.

POST-OPERATIVE CARE

• The importance of continuing these principles of care into the post-operative period is increasingly apparent.


137

Chap-09.qxd 2/2/02 1:01 PM Page 137

• Numerous studies indicate that with modern anaesthetic techniques,the occurence of intra-operative myocardial ischaemia can be limited,but that post-operative myocardial ischaemia remains common and isassociated with the development of significant cardiac morbidity andmortality.25

Post-operative care should concentrate on the maintenance of

• adequate oxygenation,

• normothermia,

• haemodynamic stability requiring the continuation of invasive cardio-vascular monitoring,

• f luid balance,

• appropriate pain control based upon regional anaesthetic or systemicopioid analgesic techniques,

• administration of anti-ischaemic medical therapy.

The appropriate location for such care is an intensive care or high dependencycare environment.

Further reading

Ramsay J. The patient with heart disease. Can J Anaesth 1996; 43: R99–107.

Foex, Howell SJ. The myocardium. Can J Anaesth 1997; 44: R67–76.

Nugent M. Anesthesia and myocardial ischemia. Anesth Analg 1992; 75: 1–3.

References

1. Department of Health. The National Service Framework for Coronary HeartDisease. Stationery Office, London, 2000.

2. Callum KG, Gray AJG, Hoile RW et al. Then and now. The 2000 report of the National Confidential Enquiry into Perioperative Deaths. Royal College ofSurgeons, London, 2000.

3. Mangano DT, Browner WS, Hollenberg M, Tateo IM. Long term cardiacprognosis following non-cardiac surgery. The study of Perioperative IschemiaResearch Group. JAMA 1992; 268: 233–9.

4. Guidelines for perioperative cardiovascular evaluation for noncardiac surgery.Report of the American College of Cardiology/American Heart AssociationTask Force on practice guidelines (Committee on Perioperative Cardio-vascular Evaluation for Noncardiac Surgery). JACC 1996; 27: 910–48.


138

Chap-09.qxd 2/2/02 1:01 PM Page 138

5. Eagle KA, Coley CM, Newell JB et al. Combining clinical and thallium data optimizes preoperative assessment of cardiac risk before major vascularsurgery. Ann Intern Med 1989; 110: 859–66.

6. Paul SD, Eagle KA, Kuntz KM et al. Concordance of preoperative clinical riskwith angiographic severity of coronary artery disease in patients undergoingvascular surgery. Circulation 1996; 94: 1561–6.

7. L’Italien GJ, Paul SD, Hendel RC et al. Development and validation of aBayesian model for perioperative cardiac risk assessment in a cohort of 1,081vascular surgical candidates. JACC 1996; 27: 779–86.

8. Guidelines for coronary angiography. Report of the American College ofCardiology/American Heart Association Task Force on assessment of diag-nostic and therapeutic cardiovascular procedures (Subcommittee on CoronaryAngiography). JACC 1987; 10: 935–50.

9. Guidelines and indications for coronary artery bypass graft surgery. Report ofthe American College of Cardiology/American Heart Association Task Forceon assessment of diagnostic and therapeutic cardiovascular procedures (Sub-committee on Coronary Artery Bypass Graft Surgery). JACC 1991; 17:543–89.

10. Guidelines for percutaneous transluminal coronary angioplasty. Report of theAmerican College of Cardiology/American Heart Association Task Force onassessment of diagnostic and therapeutic cardiovascular procedures (Subcom-mittee on Percutaneous Transluminal Coronary Angioplasty). JACC 1993;22: 2033–54.

11. Jackson G. Stable angina: drugs, angioplasty or surgery? Eur Heart J 1997; 18:B2–10.

12. Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy-1: prevention of death, myocardial infarction, andstroke by prolonged antiplatelet therapy in various categories of patients.Br Med J 1994; 308: 81–106.

13. Pfeffer MA, Braunwald E, Ciampi A et al. Effect of captopril on mortality andmorbidity in patients with left ventricular dysfunction after myocardial infarc-tion. Results of the survival and enlargement trial. N Engl J Med 1992; 327:669–77.

14. Pasternack PF, Grossi EA, Baumann FG et al. Beta blockade to decrease silentmyocardial ischaemia during peripheral vascular surgery. Am J Surg 1989; 158:113–16.


139

Chap-09.qxd 2/2/02 1:01 PM Page 139

15. Mangano DT, Layug EL,Wallace A,Tateo IM. Effect of atenolol on mortalityand cardiovascular morbidity after noncardiac surgery. N Engl J Med 1996;335: 1713–20.

16. Poldermans D, Boersma E, Bax JJ et al. The effect of bisoprolol on periopera-tive mortality and myocardial infarction in high-risk patients undergoing vascular surgery. N Engl J Med 1999; 341: 1789–94.

17. Packham MA. Role of platelets in thrombosis and haemostasis. Can J PhysiolPharmacol 1994; 72: 575–80.

18. Siu SC, Kowalchuk GJ, Welty FK et al. Intra-aortic balloon counterpulsationsupport in the high risk patient undergoing urgent noncardiac surgery. Chest1991; 99: 1342–5.

19. Shah KB, Kleinman BS, Sami H et al. Reevaluation of perioperative myo-cardial infarction in patients with prior myocardial infarction undergoingnoncardiac operations. Anesth Analg 1990; 71: 231–5.

20. Goldman L. Cardiac risk in noncardiac surgery: an update. Anesth Analg 1995;80: 810–20.

21. Association of Anaesthetists of Great Britain and Ireland. Recommendations forStandards of Monitoring during Anaesthesia and Recovery, 2000.

22. Practice guidelines for pulmonary artery catheterization. A report by theAmerican Society of Anesthesiologists Task Force on pulmonary arterycatheterization. Anesthesiology 1993; 78: 229–30.

23. Ng WS. Pathophysiological effects of tracheal intubation. In Latto IP,Vaughan RS (eds), Difficulties in Tracheal Intubation. London: WB Saunders,1997.

24. Christopherson R, Beattie C, Frank SM et al. Perioperative morbidity inpatients randomized to epidural or general anesthesia for lower extremity vas-cular surgery. Perioperative Ischemia Randomized Anesthesia Trial StudyGroup. Anesthesiology 1993; 79: 422–34.

25. Mangano DT, Browner WS, Hollenberg M et al. Associations of perioperativemyocardial ischemia with cardiac morbidity and mortality in men undergoingnoncardiac surgery. N Engl J Med 1990; 323: 1781–8.


140

Chap-09.qxd 2/2/02 1:01 PM Page 140

141

10VALVULAR HEART DISEASE AND

PULMONARY HYPERTENSION

This chapter aims to provide an outline of the management of patients with valvu-lar heart disease with or without pulmonary hypertension, who are undergoingnon-cardiac surgery:

• Not much data exists regarding valvular heart disease and perioperativerisk analysis.

• The presence of ventricular dysfunction, arrhythmia, pulmonary hyper-tension, and coexisting coronary artery disease all contribute to car-diac risk.

• Published data is conflicting; however, there is consensus that aorticstenosis (AS) is a strong independent predictor of perioperative risk.Aortic incompetence (AI) has also been associated with increased peri-operative mortality. The risks of mitral stenosis (MS) and mitral regur-gitation (MR) are less clear. It would appear that in the absence of heartfailure or recent myocardial infarction, lesions of the mitral valve do notcontribute significantly to perioperative mortality. It is, however, veryimportant to note that all of these four valve lesions are associated withan increased incidence of postoperative congestive heart failure.

In order to minimise risk to patients, all anaesthetists should undertake as meticu-lous a preoperative assessment as possible and plan the anaesthetic according toindividual patient’s diseases and the urgency and nature of the planned surgery.Anaesthetic planning should include appropriate investigations, premedicationand plans for the conduct of anaesthesia, including monitoring techniques, anal-gesic strategies and plans for the postoperative management of these patients.

Awareness of the pathophysiological processes involved in valvular heart disease andpulmonary hypertension is essential to the successful anaesthetic management of

Chap-10.qxd 2/1/02 12:07 PM Page 141

patients presenting for non-cardiac surgery. Anticipation of the effects of surgicalstresses including blood loss and other interventions such as the introduction ofpneumoperitoneum during laparoscopy, alterations in position, application oftourniquets to limbs or cross-clamps to large blood vessels and the reperfusionassociated with their release will surely facilitate in reducing morbidity and mortality in these high-risk surgical patients.

MONITORING

It is essential that appropriate monitoring be instituted before the induction of anyform of anaesthesia in these high-risk patients. In addition to routine standardmonitoring, anaesthetists should have a very low threshold for using direct arterialblood pressure monitoring. Central venous pressure (CVP) monitoring and theuse of a pulmonary artery flotation catheter (PAFC), in particular, may assist inmonitoring fluid therapy. It must be borne in mind, however, that arrhythmias canbe induced by the insertion of guide wires and catheters into the heart. Ventriculararrhythmias, in particular, are sometimes difficult to treat in patients with severeventricular hypertrophy, and a defibrillator should be present whenever such pro-cedures are being undertaken. Where the expertise is available, trans-oesophagealechocardiography (TOE) can be a very useful adjunct to monitoring, and can beused to guide volume replacement, as well as to monitor ventricular performance.Institutional facilities and expertise will inevitably vary, and monitoring should beappropriate to the ability to interpret the information obtained.

AORTIC STENOSIS

Definitions

The severity of AS is traditionally estimated at cardiac catheterisation, andexpressed as the peak-to-peak pressure gradient (PG) between the left ventricle(LV) and the aorta, PG � P (LV) � P (aorta):

• A PG greater than 50 mmHg is defined as critical AS.

It is important to remember that this value is true for patients with normal cardiacoutput, and smaller gradients may occur in those who have LV failure, and worsedegrees of AS. It is possible to quantify PGs using Doppler echocardiography,however, patients older than 50 should have coronary angiography to excludeconcomitant coronary artery disease.

Valve surface area (the area of the orifice of the open valve) is also an impor-tant measurement in AS. The normal valve surface area is 2.6–3.5 cm2. A valve surface area of 1 cm2 or less is likely to cause clinically significant aortic valveobstruction.


142

Chap-10.qxd 2/1/02 12:07 PM Page 142

Aetiology and risk

Isolated AS is usually an acquired disease. Degeneration and/or calcification mayoccur on a previously normal valve, or on a congenitally bicuspid valve. The endresult is obstruction of the aortic outflow. Isolated rheumatic disease of the aorticvalve is not common:

• Presenting symptoms include angina, syncope, and heart failure.

• Onset of symptoms from AS is usually followed by death within 2–5 yearsif the diseased valve is not replaced.

Goldman found AS to be one of nine independent predictive variables thatincreased risk of perioperative cardiovascular morbidity.1 However, O’Keefehas suggested that ‘aggressive intraoperative monitoring and prompt recognitionof haemodynamic abnormalities’ can allow for safe anaesthesia in patients withsevere, symptomatic AS undergoing non-cardiac surgery.2 Selected patients withsevere AS can therefore undergo non-cardiac surgery, but they are at greater risk,and demand appropriate monitoring and meticulous haemodynamic managementduring the perioperative time. Decisions to proceed with anaesthesia and surgeryin these patients need to be made with due consideration of severity of AS, andrelative urgency of proposed surgery. Certainly, patients with New York HeartAssociation (NYHA) class 4 symptoms, i.e. breathless at rest ought to have valvularsurgery before elective non-cardiac surgery. Patients needing urgent or emergencysurgery may benefit from valvuloplasty.

Pathophysiology

Obstructing the outflow of blood from the LV causes an increase in LV wall tension, with compensatory and characteristic concentric hypertrophy of the LV.This LV pressure overload induced hypertrophy results in reduced ventricular compliance, and higher end diastolic pressures are needed to fill the stiff LV. Patientswith AS are particularly dependent on the contribution of atrial contraction to ventricular filling,hence atrial arrhythmias can produce critical loss of cardiac outputand should be avoided at all costs.

End stage AS is associated with severe loss of compliance and the inability of theLV to sustain cardiac output, with loss of stroke volume as well as a reduced ejec-tion fraction. This leads to a state known as afterload mismatch, with the heart fail-ing because of excessive afterload, rather than contractility failure. Surgicalcorrection of the high afterload by valve replacement replacing the valve shouldrestore ejection fraction.

The hypertrophied myocardium in AS is vulnerable to ischaemia, because ofincreased oxygen demand and high wall tension. Even with normal coronary arter-ies, subendocardial ischaemia can occur, as coronary blood flow cannot keep pace

VALVULAR HEART DISEASE AND PULMONARY HYPERTENSION

143

Chap-10.qxd 2/1/02 12:07 PM Page 143

with the ventricular hypertrophy. Tachycardia should be avoided as it aggravatesischaemia.

Pharmacological afterload reduction does not alleviate the mechanical afterload tothe LV, and should be avoided as the associated reduction in diastolic blood pressuremay cause myocardial ischaemia.

Conduct of anaesthesia

It may be reasonable to consider regional anaesthetic techniques, includingepidural3–5 and even continuous spinal6 anaesthesia for selected patients with AS.However, the fall in diastolic blood pressure, and the potential bradycardia thatmay occur make these anaesthetic techniques potentially hazardous.

The choice of anaesthetic agents should be aimed at avoiding myocardial depres-sion, and avoiding peripheral arteriolar vasodilation.

Certain haemodynamic objectives need to be met or maintained:

• The patient with AS should be kept well filled, with high normal fillingpressures.

• The afterload should be maintained with judicious use of vasopressors.

• Tachycardia should be avoided.

• In particular, diastolic hypotension should be avoided, and falls in bloodpressure should be corrected with volume replacement and alpha agon-ist drugs.

• Arrhythmias should be treated promptly by cardioversion in the eventof haemodynamic compromise.

Appropriate observation and monitoring should be continued into the post-operative period.

Vasodilating and myocardial depressant induction agents (propofol and thiopen-tone) should best be avoided or used with great caution. Opiates are generally well tolerated, and may allow for reduced concentrations of inhalational anaes-thetic agents. Vecuronium, cisatracurium, and rocuronium are all suitable musclerelaxants.

AORTIC INCOMPETENCE

Definitions

Incompetence of the aortic valve results in regurgitation of a portion of the strokevolume back into the LV during diastole. The severity of AI is quantified by thevolume of regurgitant blood as estimated during angiography, or by colour flow


144

Chap-10.qxd 2/1/02 12:07 PM Page 144

Doppler echocardiography. It may be mild, moderate or severe:

• Less than 3 l/min is deemed mild, and more than 6 l/min is severe AI.

• It is possible to have regurgitant volumes in excess of 20 l/min.

Aetiology

Disease of the valve leaflets, or the wall of the aortic root, or both can cause AI.Causes include: rheumatic fever, infective endocarditis, trauma, a congenitallybicuspid valve, failure of a bioprosthetic valve, and myxomatous disease of the aorticvalve. AI can occur in the presence of ventricular septal defect, and as a consequenceof aortic dissection. More rare causes include connective tissue diseases and con-genital defects as well as treatment with the appetite suppressant phentermine–fenfluramine.

Pathophysiology

The regurgitant volume in chronic AI causes increased diastolic volume in the LV, which in turn provides a degree of haemodynamic compensation for the lossof forward stroke volume by the process of preload augmentation. Progression of the disease results in increased wall tension and eccentric hypertrophy of the LV. Theheart can be grossly enlarged. The competent mitral valve protects the left atrium(LA) and pulmonary vasculature from the volume overload of AI. In severe AI,the regurgitant jet impinges on the anterior leaflet of the mitral valve and produces the presystolic mitral murmur of Austin Flint. The LV is initially very compliant,and unlikely to become ischaemic until late in the disease process when LV failureoccurs. LV failure occurs when the chronically increased wall tension and musclemass result in loss of compliance and contractility. The onset of LV failure is followed by rapid deterioration. Reduced afterload and moderate tachycardia arethe chief mechanisms to offset the effects of AI.

A reduction in afterload as achieved by the lower diastolic aortic pressure aug-ments forward flow. A faster heart rate (�90 beats/min) reduces diastolic time and hence the regurgitant fraction.

Acute AI is poorly tolerated and patients rapidly develop heart failure with disten-sion of the LV and increased LA and pulmonary artery occlusion pressure (PAOP),as the mitral valve is unable to contain the regurgitant volume.


To minimise the effects of an incompetent aortic valve, the anaesthetic should aim to reduce or maintain a low afterload, and to keep a heart rate of about 90 beats/min. Regional anaesthesia is a logical choice where patients do not haveany other contra-indications to its use. Bradycardia should be treated aggressivelyand in low cardiac output states dobutamine or milrinone are both reasonable


145

Chap-10.qxd 2/1/02 12:07 PM Page 145

choices if inotropic drugs are needed. Moderate vasodilation from both inductionand inhalational anaesthetic agents may be beneficial to patients with AI.

MITRAL STENOSIS

Definitions

The normal mitral valve orifice in adults is between 4 and 6 cm2:

• Clinically significant MS occurs when the valve orifice is less than 2 cm2.

• When the mitral valve opening is reduced to 1 cm2, the MS is said to becritical.

The diastolic trans-mitral PG can be accurately estimated by Doppler echocardio-graphy. A gradient of

• less than 5 mmHg is consistent with mild MS,

• 5–12 mmHg is consistent with moderate MS,

• greater than 12 mmHg is severe MS.

Aetiology

Rheumatic heart disease is the most common cause of MS. Women are four timesmore likely to be affected than men. Congenital MS is rarely seen in infants andchildren. Malignant carcinoid, amyloid deposits, systemic lupus erythematosus,rheumatoid arthritis, and the Hunter Hurler mucopolysaccharidososes are otherrare causes. Large vegetations from infective endocarditis of the mitral valve, as well as LA tumours (usually myxomas) and congenital membranes in the LA (cortriatriatum) may all mimic MS.

Pathophysiology

MS causes chronic under-filling of the LV, and results in increased pressure andvolume upstream of the mitral valve. In order to generate adequate flow througha valve with a 1 cm2 orifice to maintain a normal cardiac output, the LA pressurehas to be approximately 25 mmHg. This increased LA pressure causes dilation ofthe LA and with disease progression, the pulmonary venous and capillary pressureincreases. The pressure in the pulmonary arteries (PAs) also increases and medialhypertrophy occurs in these vessels. The right ventricle (RV) has to work harderand RV hypertrophy occurs. RV dysfunction and failure is a poor prognostic signand secondary dysfunction of the right-sided valves (tricuspid and pulmonicregurgitation) occurs in severe MS.

For a given orifice size, the transvalvular PG is proportional to the square of thetransvalvular flow rate. Therefore a doubling of flow rate will quadruple the PG.


146

Chap-10.qxd 2/1/02 12:07 PM Page 146

Exercise, pregnancy, hypervolaemia, and hyperthyroidism or any other cause ofincreased cardiac output will significantly increase the transvalvular PG.

Atrial contraction contributes approximately 30% of ventricular filling in patientswith MS. Atrial fibrillation (a common feature in MS with LA enlargement),therefore significantly decreases cardiac output. Tachycardia reduces diastolic timemore than systolic time, thereby reducing the time available for flow across themitral valve. This increases the transvalvular gradient and LA pressures further.Thus, atrial fibrillation should be aggressively managed and when it does occur, itis very important to control ventricular rate.


• The single most important aspect of the anaesthetic plan in patientswith MS is to avoid tachycardia.

• Atrial fibrillation or tachycardia should be treated promptly by cardio-version or � blockade.

Filling pressures should be kept fairly high, but pulmonary oedma should beavoided. PA pressure monitoring may be desirable, as it can help to maintain opti-mal LA pressure, however, there is increased risk of PA rupture during ballooninflation, and the wedge pressure trace may not be attainable.

Afterload reduction should be avoided,as hypotension ensues with the stenotic mitralvalve precluding compensatory increase in cardiac output to maintain blood pressure.LV function is usually normal, though it will be relatively small and non-compliant.

As with AS, it is advisable to avoid vasodilating induction agents and volatile agentsshould be used with caution and titrated carefully.

MITRAL REGURGITATION

Definitions

MR may be acute or chronic. Acute MR is usually a result of infection (endo-carditis) or ischaemia with papillary muscle or chordal dysfunction. This usuallyrequires urgent cardiac surgical intervention, and anaesthesia for these patients isbeyond the scope of this chapter. Chronic MR is described as mild, moderate orsevere. The quantification of MR is made by cine-angiography, and/or colourflow Doppler and pulsed wave Doppler echocardiography.

The volume of MR can be estimated from the difference between LV stroke volumes, measured during angiography, and the effective forward stroke volumesmeasured indirectly using the Fick method:

• In severe MR, the regurgitant stroke volume approaches or evenexceeds forward stroke volume.


147

Chap-10.qxd 2/1/02 12:07 PM Page 147

Echocardiographic quantification of severity of MR is made using Doppler colourflow mapping to estimate the size and volume of the regurgitant jet, and pulsed waveDoppler to observe flow reversal in the pulmonary veins as occurs in severe MR.

Echocardiography also provides useful information regarding the cause of MRand the dimensions of the LA.

Aetiology

Abnormalities of any of the components of the mitral valve apparatus can causeMR. This includes the mitral valve leaflets, the chordae tendineae, the papillarymuscles, and the mitral annulus. Mitral leaflet pathology is usually of rheumaticorigin, although endocarditis is also implicated. The mitral valve prolapse syn-drome is another important cause. Rarely blunt or penetrating trauma may causedestruction of the mitral leaflets.

Chordal dysfunction may follow acute myocardial infarction or endocarditis.Ischaemia or infarction commonly causes posterior papillary muscle dysfunction.

Degenerative disease of the mitral annulus is common and is an important causeof MR particularly in female patients. Dilation of the annulus occurs in any cardiacdisease associated with LV dilatation. It is particularly associated with ischaemiccardiomyopathy.

Pathophysiology

The pathophysiology of MR is best thought of in terms of chronic LV overload. Theincompetent valve allows a proportion of the LV stroke volume back into the LA.The LA is highly compliant and may dilate to massive proportions. A significant pro-portion of the stroke volume will flow retrograde through the mitral valve before theaortic valve opens. The LV ejection fraction should therefore be increased (�80%).A normal ejection fraction (50–60%) may indicate depressed myocardial contractility.With large regurgitant volumes in the LA, pulmonary venous congestion ensues,and pulmonary arterial hypertension is a feature of chronic volume overload. Theregurgitant fraction is influenced by LV afterload, the size of the regurgitant defect,the PG between the LA and LV as well as the heart rate. Moderate tachycardiareduces systole and the time for regurgitant blood flow, as well as reducing diastoleand the time for diastolic filling of the LV. The absence of a competent mitral valvemeans that there is no isovolemic contraction phase during systole.


Anaesthetic goals include:

• avoid increases in afterload,

• maintain a relative tachycardia (about 90 beats/min),

• maintain relatively high filling pressures.


148

Chap-10.qxd 2/1/02 12:07 PM Page 148

As with MS, excessive filling is bad and hypotension is best treated with inotropesrather than with vasopressors. PA catheters provide useful information, both interms of pulmonary hypertension and the severity of MR, which may be seen asa ‘v’ wave on the PA wedge trace. Where inotropes are needed, both milrinoneand dobutamine are logical choices as the reduction in afterload is augmented by thereduction in PA pressure. In the event of excessive loss of systemic vascular tone,judicious vasopressor (norepinephrine) infusion is appropriate to allow continueduse of inotropes. Regional anaesthetic techniques are desirable, as the reduction in afterload, and the avoidance of the sympathetic surges associated with laryn-goscopy are beneficial to the patient. Regional anaesthetic techniques shouldnever excuse inadequate monitoring and vigilance.

PULMONARY HYPERTENSION

Definitions

Pulmonary hypertension is defined as PA systolic pressure greater than 30 mmHgand mean pressure greater than 20 mmHg. Pulmonary hypertension may beprimary (idiopathic) or secondary.

Aetiology

Primary pulmonary hypertension is a rare but progressive and fatal disease.Secondary causes of pulmonary hypertension include MR, MS, LV diseases, peri-cardial diseases, LA myxomas, congenital cardiac defects, pulmonary embolism orthrombosis, pulmonary parenchymal disease, and chronic hypoxaemic states aswell as some collagen vascular diseases.

Pathophysiology

Chronic resistance to pulmonary blood flow may cause secondary pulmonaryhypertension. Otherwise, a reactive process may cause it. Increased resistance toflow often results in an additive reactive component. The effects of pulmonaryhypertension and increased resistance to flow on the RV are complex. The RV is a thin walled structure whose function is highly influenced by its geometry.Chronically elevated PA pressures cause RV hypertrophy, which in turn results inloss of compliance and poor performance. RV failure is notoriously difficult tomanage. Apart from the deleterious effects on RV function, the hydrostatic effectsof pulmonary hypertension predispose to the development of pulmonary oedemaand hypoxaemia from ventilation perfusion mismatch. The loss of RV compliancemeans that filling pressure in the RV is difficult to optimise – under-filling leads tounderperformance while the over-filled ventricle rapidly fails. Excessive distensionof the RV also results in leftward displacement of the interventricular septum,causing a form of ‘internal tamponade’.


149

Chap-10.qxd 2/1/02 12:07 PM Page 149

One model of pulmonary circulation uses the concept of pulmonary vascularimpedance rather than that of resistance. Impedance calculations allow for the effectsof blood viscosity, pulsatile flow, reflected waves, and arterial compliance. Thus, thedynamic relationship between flow and pressure can be more comprehensivelyunderstood. Unfortunately, it remains difficult to obtain all the data required forthis calculation, and the simpler calculations of vascular resistance are used.Nonetheless, a very important factor in calculating impedance in the pulmonarycirculation is the heart rate, and for a given cardiac output, impedance is lowest ata faster heart rate, typically greater than 90 beats/min.


Severe pulmonary hypertension and incipient RV failure present some of the mostchallenging problems to the anaesthetist.

As with all anaesthetic plans, the nature of the planned surgery will influence thechoice of technique, where appropriate regional anaesthesia may be the option ofchoice:

• Increases in pulmonary vascular resistance/impedance are poorly toler-ated by the compromised RV,and hypertensive surges should be avoided.

• Hypoxaemia will worsen existing pulmonary hypertension.

• Correct fluid loading is critical, and monitoring strategies shouldinclude means to estimate both RV and LV filling pressures.

There are many therapeutic options to reduce PA pressure and support the failingor vulnerable RV:

• Increasing the inspired oxygen concentration.

• Speeding up the heart rate to relative tachycardia (90–120 beats/min).

• Vasodilators have a role to play, and nitroglycerine can reduce PA pres-sure, though often at the expense of systemic blood pressure.

• Inhaled agents such as nitric oxide and prostacycline can certainlyreduce PA pressure. Although the use of inhaled nitric oxide in particu-lar has not been universally accepted, it should be kept in mind as anoption in patients with severe pulmonary hypertension and RV failure.

Inotropic drugs may be useful:

• Milrinone has specifically beneficial effects on reducing pulmonary vascular resistance, as well as being inotropic to the RV.7

• Dobutamine also has the potential to improve RV haemodynamics andreduce PA pressure, particularly where LV failure is a feature.


150

Chap-10.qxd 2/1/02 12:07 PM Page 150

• Isoprenaline has also been used extensively in the cardiac surgicalpatient population to treat RV failure and pulmonary hypertension.

Further reading

Yarmush L. Noncardiac surgery in the patient with valvular heart disease.Anesthesiol Clin North Am 1997; 15: 69–91.

ACC/AHA guidelines for the management of patients with valvular heart disease.A report of the American College of Cardiology/American Heart Association.Task force on practice guidelines (Committee on management of patients withvalvular heart disease). J Am Coll Cardiol 1998; 32: 1486–588.

Carabello BA,Crawford FA. Valvular heart disease. N Engl J Med 1997; 337: 32–41.

Riedel B. The pathophysiology and management of perioperative pulmonaryhypertension with specific emphasis on the period following cardiac surgery.Int Anesthesiol Clin 1999; 37: 55–79.

References

1. Goldman L, Caldera DL, Nussbaum SR et al. Multifactorial index of cardiacrisk in noncardiac surgical procedures. N Engl J Med 1977; 297: 845–50.

2. O’Keefe JH, Shub C, Rettke SR. Risk of noncardiac surgical procedures inpatients with aortic stenosis. Mayo Clin Proc 1989; 64: 400–5.

3. Brian JE Jr, Seifen AB, Clark RB, Robertson DM, Quirk JG. Aortic stenosis,cesarean delivery, and epidural anesthesia. J Clin Anesth Mar–Apr 1993; 5 (2):154–7.

4. Brighouse D. Anaesthesia for caesarean section in patients with aortic stenosis:the case for regional anaesthesia. Anaesthesia Feb 1998; 53 (2): 107–9.

5. Colclough GW,Ackerman WE 3rd,Walmsley PM, Hessel EA. Epidural anes-thesia for a parturient with critical aortic stenosis. J Clin Anesth 1995; 7:264–5.

6. Collard CD, Eappen S, Lynch EP, Concepcion M. Continuous spinal anesthe-sia with invasive hemodynamic monitoring for surgical repair of the hip intwo patients with severe aortic stenosis. Anesth Analg 1995; 81: 195–8.

7. Chen EP, Bittner HB, Davis D, Van Trigt P. Milrinone improves pulmonaryhemodynamics and right ventricular function in chronic pulmonary hyper-tension. Ann Thorac Surg 1997; 63: 814–21.


151

Chap-10.qxd 2/1/02 12:07 PM Page 151

Chap-10.qxd 2/1/02 12:07 PM Page 152


11EMERGENCY ABDOMINAL AORTIC

SURGERY

Emergency aortic surgery is predominantly the surgery of leaking or rupturedaortic aneurysms, or of painful, rapidly expanding aneurysms:

• The incidence of abdominal aortic aneurysm is continuing to increase,especially in the western world1 and accounts for 2% of all deaths inmen over 60 years of age in the UK.

• The vast majority of abdominal aortic aneurysms are related to athero-sclerotic disease.

• Increasing incidence occurs with male sex, smoking, age, diabetes, andhypertension.

Associated morbidity is very common, mainly due to the common aetiologicalfeatures. Consequently there is a high incidence of coronary artery disease, reno-vascular disease, and cerebrovascular disease. Similarly, other smoking-related diseases are common. This consequently leads to a challenging operation with ahigh morbidity and mortality.

Ruptured aneurysms, untreated, are universally fatal:2

• Even with modern surgical and anaesthetic/intensive care intervention,the overall survival rates have changed little over the last few decades.3,4

• Overall mortality rates of up to 90% are reported for rupturedaneurysms5,6 in stark contrast to � 5% for elective repair.7 About halfof the deaths occur before the patient reaches hospital.6

In some cases of a ‘leaking’ (i.e. a contained bleed) or painful, rapidly expandinganeurysm, an operation better described as ‘urgent’ may be possible, but in gen-eral, patients present shocked and often moribund, and anaesthesia, operation andresuscitation are simultaneous and interrelated.

153

Chap-11.qxd 2/1/02 12:08 PM Page 153

Despite these pessimistic figures, patients who survive the peri-operative period,appear to have good long-term survival.5,8

Not surprisingly, it is difficult to conduct proper scientific studies comparing dif-ferent treatment strategies in this group of patients. There is a dearth of recentpublications, and often personal experience and evidence at an almost anecdotallevel, or that derived from, say, trauma surgery or even animal work must be extrap-olated to guide one’s approach.

It is known that results are better when both surgeon and anaesthetist are experienced and in current practice with this type of surgery. Whilst this isaccepted, it is difficult to believe that moving a moribund patient 40 or 50 milesto benefit from this expertise (should he be fortunate enough to survive the jour-ney) actually constitutes ‘best care’. The cynics would suggest that the improvedsurvival of regional centres is in part because the patients are pre-selected by having survived the ambulance journey! It seems likely, therefore, that emer-gency aortic surgery will continue to be performed in the hospital to which itpresents, perhaps by a team for whom this would not normally be a part of theirroutine work.

ASSESSMENT AND PREPARATION

Inevitably, time is limited to assess and improve these patients. As far as possible, ahistory should be sought, not so much of the presenting complaint, as of generalhealth and co-morbidity. The physiological stress of this type of surgery is verygreat, and, as with other major surgery, even patients who have been ‘very fit fortheir age’ seem to experience an acceleration of physiological age to reach theirchronological age peri-operatively. It is known that mortality increases with agebut it seems impossible to assign an absolute age of cut-off for survival. Poor prog-nostic signs include:3,6,9

• females,

• age � 75,

• actual aortic rupture,

• cardiopulmonary resuscitation (especially prior to surgery),

• hypotension (� 90 mmHg) pre- and intra-operatively (despite fluidresuscitation),

• transfusion requirements in excess of 3 l,

• raised creatinine,

• obtunded consciousness and pre-operative haemoglobin (Hb) � 10 g/dl,have also been associated with poor survival.


154

Chap-11.qxd 2/1/02 12:08 PM Page 154

To these must be added the contribution of pre-existing medical problems.Ischaemic heart disease, chronic lung disease, renal insufficiency and hypertensionare all common, and make the situation less hopeful. Heart failure is especiallyproblematic.

Ideally, the decision to proceed to surgery should be an active one, taken jointly bya senior surgeon and anaesthetist, with the accent on a possible survivor ratherthan a last desperate throw of the die.

In practice, there is often very little time for consideration or discussion, and theanaesthetist may well be presented with a ‘fait accomplis’ in that the patient andrelatives are expecting an operation, and are aware that survival is unlikely with-out. In any case, there is a large element of judgement involved, and it is natural to‘give the patient a chance’ unless it is obvious (usually only in hindsight) that thepatient has no realistic prospect of survival. There is often very little to go on inthe way of investigations:

• Measurement of full blood count, urea and electrolytes, and blood glu-cose can all guide further management.

• A 12-lead ECG can give an indication of previous cardiac insults as wellas present cardiac function/rhythm.

• Plain chest X-rays offer little useful information in these patients andcan delay surgery.

• Blood (10 units), fresh-frozen plasma (FFP), and platelets should berequested.

• If time permits, and a peripheral pulse is palpable, an arterial blood gasanalysis may give helpful information as to the true severity of thepatient’s condition, and, of course, it is helpful if an arterial cannula isplaced to obtain the sample.

Fluid resuscitation in its own right is also controversial.10 There is clear evidencethat over zealous raising of blood pressure prior to surgery may lead to furtherbleeding, dislodging of haematoma and dilutional coagulopathy; in effect making matters worse. Against this is the problem of prolonged tissue ischaemiaexacerbating reperfusion injuries, and renal and cardiac ischaemia. (There are similar issues involving patients with penetrating trauma which are discussed inChapter 6.)

The controversy as to the relative benefits of colloid or crystalloid in resuscitationremains unresolved after many years of investigation. This surely means that, pro-vided the different dynamics of the two are understood, and allowed for, it doesnot really matter. This issue is further discussed in Chapter 6.

EMERGENCY ABDOMINAL AORTIC SURGERY

155

Chap-11.qxd 2/1/02 12:08 PM Page 155

CONDUCT OF ANAESTHESIA

Once the decision to operate has been taken, the patient is transferred to theatre,where resuscitation can continue simultaneous with surgical preparations. Thesepatients are often distressed, and may be in considerable pain. Clearly, to be effect-ive, analgesics must be given intravenously, but great care must be taken, as thepatient will be exquisitely sensitive to their depressant effects. They will be relianton abdominal tone to tamponade the aneurysm and sympathetic nervous systemactivity to maintain their blood pressure:

• Anaesthetists should be clear that getting the aorta cross clamped iswhat is going to save the patient’s life.

• In a shocked patient surgery should not be delayed whilst central linesand arterial lines are placed.

• At least two wide-bore (16g or 14g) cannulae are the minimum pre-induction.

• Fluid warmers and rapid infusion devices should be available andprimed.

At least two anaesthetists are required in the initial stages, allowing one to con-centrate on anaesthesia and respond to the patient’s rapidly changing physiologyand the second to perform practical tasks such as invasive line placement when itis deemed safe to do so. At least one of the anaesthetists should be a consultant.

The patient is resuscitated and anaesthetised on the operating table. Access toradial arteries and peripheral veins will be required so the patient’s arms are placedout on arm boards. Despite the urgency of the operation, attention still needs tobe made to pressure area care, and vulnerable nerves. A urinary catheter should bepassed, if this has not been done already, and the hourly measuring chamberbrought to the head of the table, so that it can be observed easily by the anaes-thetist. At some appropriate stage a nasogastric tube should be passed as theretroperitoneal haematoma invariably causes a post-operative ileus:

• Monitoring with ECG, SpO2, NIBP is the mandatory minimum priorto induction, with gas analysis (capnography, O2, volatile agent) andventilatory parameters after induction.

• When practical, core temperature, invasive arterial pressures, centralvenous pressures and possibly pulmonary artery pressure monitoringwill be required.

• Resuscitation drugs including epinephrine, vasoconstrictor agents (directacting alpha agonist) such as metaraminol or methoxamine should beavailable and prepared for use.


156

Chap-11.qxd 2/1/02 12:08 PM Page 156

• The surgeon should be scrubbed, with ‘knife in hand’ and the patient’sabdomen both ‘prepped’ and draped prior to induction.

Temperature maintenance is important and although cooling may be said to givesome protection to the vessel rich organs, the effects of hypothermia on bloodclotting and metabolism more than outweigh this, so an active warm air heatingblanket is placed over the patient’s torso, arms and head.

Choice of anaesthetic will be a personal one (see also Chapter 6 for discussions onchoice of anaesthetic agent in the shocked patient) and in reality it may be the carewith which it is given that is most important:

• The guiding principle however, is to use cardiovascularly stable drugsthat will produce the least impact on an already severely compromisedpatient.

• A ‘rapid sequence’ technique with cricoid pressure is mandatory.The patient may not be fasted, and even if he is, there is clearly intra-abdominal pathology that will impair gastric emptying!

• Various combinations of drugs can be used including midazolam,etomidate, ketamine, thiopentone, fentanyl or remifentanil.

• This is followed by an intubating dose of suxamethonium and intubation.

Surgery usually commences as soon as intubation is achieved. Clearly communi-cation between anaesthetist and surgeon is essential throughout the operation, butnever more so than at this point.

Maintenance of anaesthesia again follows the same principles as induction:

• Cardiostability is the primary requirement.

• Again, combinations of opiates, volatile, and benzodiazepines are fre-quently used together with a non-depolarising neuromuscular blockingagent.

• Ventilation with oxygen/air mixtures avoids the cardio-depressanteffects of nitrous oxide on the circulation, helps prevent peripheralatelectasis, and expansion of gas spaces in the bowel. An FiO2 of 0.5 isprobably the minimum that is appropriate. The patient should be ventilated to normocapnia. This will be influenced by arterial blood gasanalysis.

• Given the unpredictability of the coagulation process peri-operatively,and the de-stabilisation of the cardiovascular picture likely to be caused,the use of epidural regional blockade in the emergency context is not wise.


157

Chap-11.qxd 2/1/02 12:08 PM Page 157

With the onset of anaesthesia and loss of abdominal wall tamponade, there willusually be a steep fall in blood pressure. This can be opposed by the use of fluidsand vasopressors, but at this stage, the patient relies upon the rapid and effectiveplacing of the cross-clamp to give control of the aortic bleed:

• This involves a laparotomy using either a transverse abdominal or vertical midline incision.

• The aorta and haematoma are then identified in the posterior peri-toneum and the superior neck of the aneurysm clamped.

• Surgical mishaps such as aortic, or worse, caval tears can lead to a rapiddemise of the patient, due to the torrential bleeding that ensures.

• Once the aorta is clamped and bleeding controlled, the blood pressureshould start to come up with continued resuscitation, and the operationmoves into its middle phase.

• This involves the surgeon opening or resecting the aneurysm, evacuat-ing haematoma and by-passing the aneurysm with an artificial graft.

CROSS-CLAMP PHASE

The magnitude and significance of the haemodynamic changes caused by crossclamping are related to the level of the aorta at which the clamp is applied.11 Thereduction in effective vascular capacity causes:

• ↑ Afterload.

• ↑ Preload and pulmonary capillary wedge pressure (PCWP).

• ↑ Myocardial oxygen demand. Myocardial ischaemia is common whichmay respond to GTN.

• Cardiac output often falls especially in patients with coronary arterydisease.

• Some of these changes, e.g. the increase in afterload may be controlledwith either volatile agents or vasodilators.

• These changes may be reduced in patients who are hypovolaemic –thus the above effects may be less significant in emergency, bleedingpatients than in elective aortic surgery.

High aortic clamps may result in spinal cord ischaemia. Blood supply to the thoracolumbar area of the cord is derived from the artery of Adamkiewicz – vulnerable to the ‘steal’ phenomenon. Prevention of this disastrous complicationis helped by fast surgery and maintaining best possible cardiac function.


158

Chap-11.qxd 2/1/02 12:08 PM Page 158

Renal blood flow falls with aortic cross clamping – 80% fall with suprarenalclamping but even infrarenal clamping causes falls of approximately 40%. There isno reliable way to preserve renal function (see also the chapter on PerioperativeRenal Insufficiency and Failure). A short cross-clamp time is crucial but changesin renal blood flow and renal vascular resistance may persist for some time.

The cross-clamp phase usually allows the anaesthetist to stabilise the patient andprepare them for the reperfusion stage of the operation.

Although the retro- or intra-peritoneal bleed initially leads to a period of hyper-coagulability,by the time they reach theatre, the patient will invariably have a coagu-lopathy, in part consumptive, in part dilutional. At least 4 units of FFP and plateletsearly on in this stage are usually required, ideally aided by coagulation studies. Thehelp and understanding of the blood bank is crucial to ultimate success, and theymay be able to offer guidance as to appropriate component therapy. Some cautionis required, because each coagulation study is a snapshot of a rapidly changing situ-ation, which together with the ‘lead time’ in transporting and analysing the speci-mens, may require some imagination to interpret as the case unfolds. As well asguiding red cell transfusion therapy, the availability to measure Hb in theatre ismuch more reactive, and a useful aid to interpreting other lab results:

• Arterial and central venous access should now be achieved, if notalready in place.

• Arterial blood gases analysis can give an indication of the hypoxic insultalready sustained (there is often a gross metabolic acidosis). It should beunderstood that this represents a measure of the severity of the patient’spredicament rather than a simple indication for, say, bicarbonate ther-apy. The aim is to improve global perfusion so that the figures improverather than to treat the figures themselves. The situation is often muchworse than expected.

• The use of a pulmonary artery flotation catheter (PAFC) is still contro-versial, and many would argue that the use of a properly transduced cen-tral venous line should give adequate filling pressure trend information.

• Should information on cardiac output be required, a PAFC may be themost generally available method, but the more widespread availabilityof non-invasive methods may ultimately prove more useable. In truth,with such an abnormal cardiovascular system, the interpretation ofderived information is difficult in any case.

Unclamping the aorta

In the ideal situation, the patient will have adequate cardiac filling pressures, a reasonable arterial blood pressure (compared to their normal), be warm, perfusing


159

Chap-11.qxd 2/1/02 12:08 PM Page 159

their periphery, have some urine output and have an acid/base status returningtowards normal.

In reality, the ideal situation is rarely even approached, and the process of remov-ing the clamp (particularly for a straight graft) may well be difficult. Again,co-ordination and co-operation between surgical and anaesthetic teams is crucial,with adequate warning to the anaesthetist of the intention to remove the cross-clamp, and the willingness to do so progressively, or even repeatedly to re-clampthe aorta to minimise the incremental physiological effect.

The blood pressure will fall following reperfusion. In an elective case, it is realisticto aim to limit this to perhaps 20 mmHg decline in systolic pressure. In the emer-gency operation, falls are likely to be much greater. There are, broadly, three causesfor this:

• Increased capacity/reduction in systemic vascular resistance leads, ineffect, to central relative hypovolaemia. This is exacerbated by anyactual hypovolaemia.

• As the tissues are reperfused, reactive hyperaemia occurs, which reducesSVR further. Several hours worth of metabolic/ischaemic products arethen washed out of the stagnant lower body and returned to the circu-lation. Metabolic acidosis results. (Bicarbonate used to be given at thispoint but does not reliably prevent the falls in blood pressure.)

• This leads to an immediate fall in myocardial contractility, and a dispro-portionate fall in cardiac output. These products are also intimatelyrelated to the reperfusion injury, which we will discuss below.

InotropesThe main object of inotropic support in abdominal aortic surgery is to optimiseorgan perfusion in the heart, kidney, brain and gut. If poor arterial blood pressurepersists after volume correction, then some form of cardiac output monitoringwill be required to guide therapy:

• Dopexamine, with its preferential improvement in splanchnic perfusionand inotropic effects may have a role. Unfortunately, this agent also pro-duces vasodilatation, and the dose may be limited in the dynamic situ-ation by falls in systolic pressure, requiring norepinephrine to offset this.It is probably too early to be sure of the role of this agent.

• Dopamine and dobutamine have both been used but tend to promotea tachycardia, and the improvement in cardiovascular variables may notbe reflected by improved tissue perfusion.


160

Chap-11.qxd 2/1/02 12:08 PM Page 160

TRANSFUSION ISSUES

If a ‘cell saver’ is used for elective aortic surgery many patients will require noautologous blood at all (although the average ‘transfusion’ will be between 2 and 4 units). In emergency surgery, the situation is different, and the patient will oftenstart the operation with a low Hb (absolute loss into the retroperitonium or theabdomen, and then by dilution as the resuscitation continues). As mentionedabove, a consumptive and dilutional coagulopathy is the usual, and should betreated aggressively.

Aortic surgery is perhaps the area to which homologous blood collection and ‘salvage’ (by ‘cell saver’, or similar device) is best suited, and it is possible to avoidtransfusing large volumes of ‘bank blood’ by use of these systems. The productfrom a salvage device is red cells in saline with a variable, but usually higher thannormal haematocrit, suspended in crystalloid. Although helpful in ensuring a rea-sonable number of well-functioning red cells in the circulation, all of the plasmasalvaged is lost, and the coagulopathy is likely worse than measurements of Hb orhaematocrit would suggest:

• The use of clot-enhancing alginate precludes the further salvage ofcells, so the timing of its first use is important.

We aim for a Hb concentration between 8.5 and 10 g/dl, since this allows reason-able oxygen content, whilst the reduction in viscosity can actually improve tissuedelivery of oxygen.

REPERFUSION INJURY AND ORGAN PROTECTION

It should be understood that, even if the operation to repair an aortic aneurysmwere not to result in any periods of hypotension, the vast majority of patients willhave suffered a significant ischaemic injury before reaching hospital, and eachhypotensive stress after this serves to compound the problem. Most anaesthetistswould recognise that patients become progressively more difficult to resuscitatefrom each hypotensive episode, ultimately becoming refractory to all attempts(what used to be called irreversible shock):

• The ischaemic/reperfusion injury is central to this phenomenon.

• Reduced systemic perfusion due to the hypovolaemic shock and therelatively ischaemic lower part of the body all have an effect.

When the ischaemic tissues are reperfused an increasingly complex range of chemico-humoral reactions take place. This is some what out of the scope of this chapter but an excellent review by Gelman11 is worth reading. Products


161

Chap-11.qxd 2/1/02 12:08 PM Page 161

of ischaemic/anaerobic metabolism and oxygen metabolism are released into thecirculation including:

• hypoxanthine,

• oxygen free radicals,

• prostaglandins.

Micro-aggregates from the legs, endotoxins from the gut, large fluxes in the sym-pathetic nervous system plus the effects of neutrophils and complement activationall lead to tissue damage both immediately and in the longer term. The systemicinflammatory response syndrome (SIRS) is common. The sequelae of this willpersist for many days into the post-operative period, and may ultimately comprom-ise the patient’s recovery.

Despite close study, there seem to be very few options to ameliorate this effect.In animal work, hypoxic reperfusion (i.e. reperfusing the ischaemic area withblood or clear fluid having a low oxygen content so that metabolites are clearedprior to re-oxygenation) seems to offer some benefit, but it is difficult to see howthis might be achieved clinically. An alternative strategy is to attempt to ‘scavenge’these harmful products before they inflict too much damage.

Mannitol:

• inhibits the ischaemia-induced neutrophil oxidative activity and conse-quent hyperperoxide production,12

• acts as a free radical scavenger,

• decreases arachidonic acid breakdown,

• helps to promote a diuresis by osmotic action.

Mannitol (0.2–0.5 g/kg), prior to reperfusion is thus frequently given. Otherstrategies including non-steroidal anti-inflammatory drugs (NSAIDs), allopurinol,heparin and N-acetylcysteine13 have all been advocated at times. However, it isclear that no single metabolic pathway is exclusively responsible for reperfusioninjury, and this is likely to account for the poor performance of some of theseinhibitors in the emergency situation. In recent years, the role of activated neutrophils has undergone close scrutiny14 and may ultimately result in therapeuticprogress.

AFTER CARE AND ANALGESIA

• Provided the patient survives the operation, emergency patients requirea period of physiological support, which can only be realistically givenin an intensive care unit. Elective cases may be suitable for a highdependency unit.


162

Chap-11.qxd 2/1/02 12:08 PM Page 162

• Although we aim to have our elective cases breathing spontaneously atthe end of the operation, we would continue controlled ventilation inthe emergency group.

The immediate post-operative period is typified by cardiovascular instability, hypo-thermia, risk of re-bleeding and considerable physiological disturbance. Multi-system support is frequently required due to SIRS and multiple organ failure.Patients of this age, and with the typical levels of co-existing disease that theyexhibit, have little in the way of reserve, and unless they show a rapid improvementover the first and second post-operative days, tend to enter a downward spiral ofworsening SIRS from which they cannot recover.

LATE MORTALITY

• It is disappointing that, despite considerable improvement in our under-standing of the processes at work when an aortic aneurysm ruptures,and in the quality of the care we can offer these patients, the overallmortality for the condition remains stubbornly high.

• Death on the table, particularly following the induction of anaesthesia,is now uncommon, but this seems to have been converted into latemortality from multi-organ failure rather than into ultimate survival.

• It is very difficult to predict outcome, and we have all been surprised atpatients who have survived against our expectations, as well as disap-pointed by those who have succumbed despite our (always guarded!!)optimism. It seems likely that the ultimate key to our management ofthis condition will lie in further understanding, and better control ofthe ischaemia/reperfusion injury and prevention of rupture by screen-ing and early elective surgery.

Further reading

Thomson DA, Gelman S. Anesthesia for major vascular surgery. Clin Anesthesiol(Bailliere’s best practice and research) 2000; 14: 1–235.

References

1. Fowkes FG, Macintyre CC, Ruckley CV. Increasing incidence of aorticaneurysms in England and Wales. Br Med J 1989; 298: 33–5.

2. Macgregor JC. Unoperated ruptured abdominal aortic aneurysm: aretrospective clinico-pathological study over a 10-year period. Br J Surg 1976;63: 113–16.


163

Chap-11.qxd 2/1/02 12:08 PM Page 163

3. Sasaki S, Sakuma M, Samejima M et al. Ruptured abdominal aortic aneurysm:analysis of factors influencing surgical results in 184 patients. J Cardiovasc Surg1999; 40 (3): 401–5.

4. Rutledge R, Oller DW, Meyer AA et al. A state-wide, population-based timeseries analysis of the outcome of ruptured abdominal aortic aneurysms. AnnSurg 1996; 223: 492–502.

5. Milner Q JW,Burchett KR. Long term survival following emergency abdom-inal aortic aneurysm repair. Anaesthesia 2000; 55: 432–5.

6. Semmens JB, Norman PE, Lawrence-Brown MM et al. Influence of genderon outcome from ruptured abdominal aortic aneurysm. Br J Surg 2000; 87(2): 191– 4.

7. Scott RA, Wilson NM, Ashton HA et al. Influence of screening on the inci-dence of ruptured abdominal aortic aneurysm: 5 year results of a randomisedcontrolled study. Br J Surg 1995; 82: 1066–70.

8. Hiatt JCG, Barker WF, Machleder HI et al. Determinants of failure in thetreatment of ruptured abdominal aortic aneursym. Arch Surg 1984; 119:1264 –8.

9. Urwin SC, Ridley SA. Prognostic indicators following emergency aorticaneurysm repair. Anaesthesia 1999; 54: 739–44.

10. Brimacombe J, Berry A. A review of anaesthesia for ruptured aortic aneurysmwith special emphasis on preclamping fluid resuscitation. Anaesth Inten Care1993; 21: 311–23.

11. Gelman S. The pathophysiology of aortic cross-clamping and unclamping.Anaesthesiology 1995; 82: 1026–60.

12. Paterson I, Klausner J, Pugatch R et al. Non-cardiogenic pulmonary oedemaafter abdominal aortic aneurysm surgery. Ann Surg 1989; 209: 231– 6.

13. Kretzschmar M, Klein U, Palutke M et al. Reduction of ischaemia-reperfusionsyndrome after abdominal aortic aneurysmectomy by N-acetylcysteine butnot mannitol. Acta Anaesthesiol Scand 1996; 40 (6): 657–64.

14. Welbourn CR, Goldman G, Paterson IS et al. Pathophysiology of ischaemiareperfusion injury: central role of the neutrophil. Br J Surg 1991; 78: 651–5.


164

Chap-11.qxd 2/1/02 12:08 PM Page 164

165

12GASTROINTESTINAL SURGERY

Gastrointestinal surgery is often high risk surgery. Anaesthesia for this surgery is asubject that is not well covered in many books published on the practice of anaes-thesia. For example, the 4th edition of Anesthesia, the highly respected work editedby Miller, has no section devoted to Gastrointestinal Anaesthesia in the section onSubspeciality management. This is surprising considering gastrointestinal surgerymakes up a large part of daily practice in district general and teaching hospitalsalike. It seems to be assumed that knowledge of providing anaesthesia for the highrisk gastrointestinal patient will be gleaned purely from experience gained in man-aging other patients. In other words, anaesthesia for gastrointestinal surgery is just‘General Anaesthesia’.

Paradoxically, certain rare conditions encountered in surgery in the abdomen,e.g. carcinoid or pheochromocytoma are well covered in standard texts and willnot be covered here. Similarly, management of conditions such as acute pancreatitis,though surgical, are not commonly operated upon in most centres and are wellcovered in intensive care unit (ICU) textbooks. These conditions will also not bediscussed in this chapter.

GASTROINTESTINAL SURGERY – THE ULTIMATE IN HIGH RISK

The general public (and many practitioners) would no doubt consider such sur-gery as open heart surgery as being amongst the most riskiest of surgical oper-ations in terms of immediate and early mortality. In fact, certain relatively commongastrointestinal operations are arguably amongst the highest risk procedures per-formed. For example,perusal of a recent standard surgical text1 reveals the expected

Chap-12.qxd 2/2/02 1:05 PM Page 165

operative mortality for the following operations and conditions:

Operation Operative mortality (%)

Ca colon resection 5–10Large bowel obstruction 10Small bowel obstruction 30Ca pancreas 20Ca oesophagus 10

Many of these expected mortalities will be increased if performed as an emergency.

The lessons to be learnt are:

• Gastrointestinal surgery is high risk surgery and warrants senior inputand appropriate facilities.

• With the expected mortality, palliation may be better than attempting acure in some patients.

• The examples of good practice available from CEPOD, discussed in anearlier chapter, must be learned.

Reasons for being high risk

• Coexisting medical diseases. Many of the patients are elderly with sig-nificant medical problems.

• Type of surgery. Often long procedures with significant blood loss, fluidshifts, electrolyte and nutritional problems and significant post-operativepain.

• Abdominal surgery is associated with a profound physiological stressresponse.

• Emergency or elective. Many of these patients will present as urgent oremergent cases. This is well recognised to be associated with a worseoutcome. Problems associated with emergency cases include less timeto evaluate, investigate and treat patients.

• High incidence of hypovolaemia.

• Abdominal surgery is associated with significant respiratory embarrass-ment – upper abdominal more than lower.

• Many patients will suffer from pre-, peri- or post-operative sepsis. Weall have a lethal dose of endotoxin contained within our gut!


166

Chap-12.qxd 2/2/02 1:05 PM Page 166

GENERAL PRINCIPLES OF INTRAOPERATIVE MANAGEMENT

• Patient positioning important for surgical access for certain incisions.One should be guided by the surgeon but should not forget our respon-sibilities for protecting skin, joints and nerve function.

• Hypothermia is common during major intra-abdominal surgery and is directly related to the length of the procedure. Heat loss is maximalduring the time that the peritoneum is open. Warmed anaesthetic gases, IV fluids and forced air warming may all be required to maintainbody temperature. The adverse effects of perioperative hypothermia are discussed in the chapter on the critically ill patient in the operating theatre.

• Monitoring should be appropriate to the status of the patient. Thereshould be a low threshold for invasive monitoring for high risk patientsundergoing gastrointestinal surgery. Large fluid losses may occur includ-ing post-operative 3rd space losses (see below). CVP monitoring maybe useful to guide fluid requirements after surgery.

• Large bore IV access may be required.

• Prophylactic antibiotics are required. Single dose prophylaxis may bepreferable to multiple doses.

FULL STOMACH/ASPIRATION RISK

The management of a patient with a full stomach is an ongoing matter of debatein anaesthetic texts.

Current Starvation Guidelines for adults revolve around 6 h fasting from solids and2–4 h fasting from clear liquids. In the gastrointestinal setting, many patients mustbe considered to have ‘full stomachs’:

• Emergency surgery. Pain, stress, trauma, opioids, abdominal pathologycan all decrease gastric emptying. Patients are also likely to be unstarved.

• Many emergency patients may be vomiting.

• Hiatus hernia confers a recognised risk of aspiration.

• Bowel obstruction obviously is a potent source of pulmonary aspirationof gastrointestinal contents. Faecal matter is a source of severe problemsif inhaled.

• All cases of peritonitis are associated with delayed gastric emptying.

GASTROINTESTINAL SURGERY

167

Chap-12.qxd 2/2/02 1:05 PM Page 167

Prevention is well established and includes:

• Nasogastric (NG) tube: Used pre-induction to empty stomach. Need notbe removed prior to induction but should be left vented to air.

• Antacids: Consider use of non-particulate antacids. Can increase gastricvolume.

• Histamine receptor antagonists: Effective at reducing both gastric volumeand acidity but relatively slow onset of action.

• Prokinetic agents: Intravenous Metoclopramide will decrease gastric volume in 15 min. Prokinetic agents are contraindicated in the presenceof intestinal obstruction.

• Proton pump blockers: A single dose of Omeprazole given the nightbefore surgery is inadequate but more frequent doses pre-operativelymay be effective.

Although prevention and prophylaxis are important the mainstay of prevention ofaspiration is appropriate anaesthetic technique including:

• Identification of patients with difficult airways.

• Avoidance of prolonged or unnecessary airway manipulations.

• Awake fibreoptic intubation is always an alternative but has not achievedwidespread popularity in the UK.

• Cricoid pressure. Properly applied cricoid pressure is the mainstay of protection against aspiration during induction of anaesthesia. Care shouldbe taken as improperly applied pressure may obstruct laryngoscopy.

ANAESTHETIC FACTORS

There are three main areas of controversy

1. Nitrous oxide. Nitrous oxide is highly soluble – 34 times as soluble as nitro-gen. Thus during anaesthesia nitrous oxide rapidly enters gas filled spacesincluding the bowel. This may cause problems with bowel distension and,after prolonged surgery, restrict abdominal closure and contribute to intra-abdominal hypertension. In obstruction, the increase in intraluminal pressurecould precipitate perforation. However, in the only study to examine thisissue patients receiving nitrous oxide had no increase in bowel distension orpost-operative bowel function.2

The increase in intraluminal gas is said to increase the incidence of post-operative nausea and vomiting but studies show conflicting results.


168

Chap-12.qxd 2/2/02 1:05 PM Page 168

2. Epidural analgesia. Post-operative ileus involves sympathetic and parasym-pathetic pathways. These can be blocked by epidural analgesia leading to areduction in the incidence of ileus following abdominal surgery.3 Thebowel is relatively contracted – which some surgeons dislike.

Post-operative nitrogen balance after bowel surgery is improved by extraduralanaesthesia.4 Other benefits of epidurals are discussed in the chapter onregional anaesthesia.

3. Opioids. Morphine has a major effect on bowel motility, significantly pro-longing the time for recovery of bowel function after colonic surgery.5

Pethidine is said to cause less spasm of the sphincter of Oddi than morphineso may be preferred in biliary colic.

The anaesthetist and the anastomosis

Surgeons are often concerned that the increase in intestinal motility with epi-durals may increase the incidence of anastomotic breakdown. However, a reviewof 12 trials has found no evidence of harmful effect6 (but concluded that largerstudies are needed for a definitive answer). In fact, by increasing intestinal bloodflow one might expect epidural anaesthesia to have a favourable effect on a bowelanastomosis. In retrospective studies in Australia, death rates and anastomotic leak-ages were less in groups of patients receiving post-operative epidural infusions oflocal anaesthetics.7

Neostigmine increases intraluminal pressure (not prevented by atropine) and hasbeen implicated in the past (mainly by surgeons) as a cause of anastomotic break-down. Animal studies do not support this assumption. A large patient study foundno difference in the rate of anastomotic leakage with or without neostigmine.8

Surgical factors are undoubtedly more important than anaesthetic factors indetermining the fate of the anastomosis:

• The site of anastomosis is important – 25% leakage in low anteriorresection anastomoses.

• The skill and experience of the surgeon are crucial.

• Tension on the anastomosis.

• Intra-abdominal infection is a major factor.

• Preservation of blood supply to the gut.

• No difference between staples and hand sutured anastomoses.9

Anaesthetists can help by maintaining good oxygenation (including into the post-operative period), prompt treatment of hypovolaemia and hypotension and avoid-ing hypocapnia.


169

Chap-12.qxd 2/2/02 1:05 PM Page 169

For the last word on this subject, I quote from a recent surgical standard text:1

Anastomotic dehiscence is related to two main factors: the level of the anastomosis and the surgeon.

Neither of these factors is within the control of the anaesthetist.

HYDRATION/FLUID THERAPY

Patients for major gastrointestinal surgery may well have a large fluid deficit pre-operatively and this should be assessed and treated before bringing the patient totheatre unless surgery is an emergency. If this is the case fluid resuscitation shouldoccur in the anaesthetic room preferably prior to induction. Causes of hypo-volaemia:

• Emergency operation: May have long starvation time secondary topathology.

• Bowel preparation: Can cause large fluid and electrolyte losses.

• Vomiting.

• Diarrhoea.

• Fistulae losses.

• ‘3rd Space’ losses. Intraoperative handling of the tissues and bowel causetissue oedema and fluid sequestration within the bowel. As much as8 ml/kg/h of fluid may need to be infused intraoperatively to coverthese losses. Extra fluids will also be required after surgery. The boweloedema will contribute to intra-abdominal hypertension and compart-ment syndrome as discussed in the chapter on Perioperative RenalInsufficiency and Failure.

Volume loading is of benefit to gut blood flow – which is important for integrityof bowel anastomoses. One study clearly showed that fluid loading after induc-tion of anaesthesia to a maximum Stroke Volume led to a reduction of the incidenceof low pHi (an indirect indicator of gut blood flow) from 50% to 10%.10

Perioperative blood transfusion

It has been widely suggested that transfusion of stored red blood cells results inimmune modulation and an increase in infective complications and, even moreworryingly, an increase in cancer recurrence rates. However, a large prospectivetrial using leukocyte depleted blood has failed to find any association betweentransfusion and tumour recurrence rates. However, there was still an increase ininfection rates and a worse outcome.11


170

Chap-12.qxd 2/2/02 1:05 PM Page 170

Thus, although the recurrence of the cancer may not be a significant concernthere are still good reasons to limit, where possible, the transfusion of stored redblood cells. This is further dicussed in Chapter 15. Approaches to reduce peri-operative blood loss to limit the requirement for transfused blood are beyond the scope of this chapter.

NUTRITION AND NUTRITIONAL SUPPORT

Nutrition is extremely important in the surgical high risk patient. In its absencethere will be:

• muscle atrophy and weakness,

• decrease in immune function,

• decrease in wound healing,

• impaired cough and respiratory function,

• gut mucosal atrophy with possible increased gut wall permeability.

The post-operative patient may be markedly catabolic with increased nitrogenlosses. Many patients will also be malnourished pre-operatively especially patientswith dysphagia or vomiting and patients with a catabolic illness such as malig-nancy or inflammatory bowel disease.

In malnourished patients pre-operative nutritional support would intuitively seem important but controlled trials have not demonstrated the expected benefits.Post-operative nutritional support is vital to minimise the negative nitrogen balance and to avoid the complications of malnutrition listed above. The enteralroute is to be preferred.

Total parenteral nutrition

• Only indicated if unable to feed the patient via the enteral route.

• More expensive.

• More metabolic problems and associated with a reduced survival due toinfectious complications in cancer patients receiving chemotherapy.

• Requires dedicated central venous access (with all its complications).

• Incidence of infectious complications related to the care of the siterather than the type of site e.g. tunnelled versus non-tunnelled.

• Usually adminstered as a 2.5 or 3 L ‘big bag’ feed with all the nutri-tional components mixed in pharmacy in a laminar flow cabinet understrict aseptic conditions.


171

Chap-12.qxd 2/2/02 1:05 PM Page 171

A large study demonstrated an increase in infectious complications in patientsundergoing major surgery who received perioperative total parenteral nutri-tion (TPN).12 Guidelines suggest that pre-operative TPN is not appropriate inpatients with only mild to moderate degrees of pre-operative malnutrition.13

Post-operative TPN is only required if the patient cannot receive enteral nutritionwithin 7–10 days.

The most important conclusion is that short-term TPN is probably to be avoidedbecause the benefits do not warrant the complications.

Enteral nutrition

• Supports normal gut flora.

• Possibly reduces gastrointestinal bleeding from stress ulcers.

• Decreases infectious complications in surgical patients.

• May preserve the gut barrier and prevent bacterial or endotoxintranslocation.

• Reduced incidence of acalculous cholecystitis.

• Cheaper than TPN and with less metabolic and infectious complications.

Early enteral nutrition (EN) is clearly associated with improved outcome14

(though a minority view would hold that EN is only better because TPN is worse,i.e. EN avoids the metabolic and infectious complications of TPN).

Certain new nutritional supplements may be of worth:

• Glutamine is an essential fuel for bowel mucosal cells (may protectbowel mucosa). Supplementation has been shown to lessen the negativenitrogen balance after major surgery.15 A recent small study in ICUpatients found a significant improvement in outcome in those patientswho received glutamine supplements.16

• The idea that certain nutrients, e.g. Omega 3 fatty acids, nucleotidesand, arginine in a feed (Impact) may directly stimulate the immune system is extremely exciting. Several studies of major gastrointestinalsurgery have found a reduced length of stay on average, a decrease ininfectious complications and less overall costs (despite the compara-tively greater cost of the feed) in the group fed with this preparation.17

RESPIRATORY ASPECTS OF ABDOMINAL SURGERY

Respiratory function is significantly impaired after abdominal surgery, especiallyupper abdominal surgery. A combination of factors are involved:

• Reduced FRC.


172

Chap-12.qxd 2/2/02 1:05 PM Page 172

• Pain leading to decreased cough and atelectasis. Pain is greater followingupper abdominal incisions compared with lower abdominal incisions.

• Diaphragmatic dysfunction.

Patient factors may increase the incidence and severity of post-operative respira-tory complications:

• Smoking.

• Obesity.

• Chronic obstructive pulmonary disease.

Anaesthetic factors:

• Most studies have examined the effects of different analgesic regimens.Meta analysis confirms that the excellent post-operative analgesia with continuous epidural analgesia leads to a reduction in respiratorycomplications.18 The impairment in respiratory function followingabdominal surgery is lessened – but respiratory function is still reducedcompared to pre-operative values.

• Large tidal volumes during anaesthesia are beneficial on respira-tory function19 but it is less certain if there are residual benefits post-operatively. PEEP has been shown to be beneficial in morbidly obesepatients but not normal patients.20

• Interestingly, pancuronium is associated with post-operative complica-tions when given for patients undergoing lengthy operations comparedwith atracurium21 – implying perhaps that repeated doses of pancuro-nium are associated with residual neuromuscular blockade.

Surgical factors:

• Surgical factors have been less well studied but it seems that length ofsurgery and blood loss are predictors of post-operative respiratory com-plications.

• Intra-abdominal sepsis leads pre-operatively to a classical rapid shallowbreathing pattern and increased ventilatory demand due to increasedenergy expenditure with potential for post-operative problems espe-cially if the patient’s reserve is such that they cannot meet the increaseddemands.22

• A recent study has challenged some assumptions regarding the site ofincision as a factor in respiratory complications. The results showed no statistically significant difference in pulmonary function, morbidityor analgesia consumption following transverse or midline abdominalincisions.23


173

Chap-12.qxd 2/2/02 1:05 PM Page 173

Several randomised trials have demonstrated the role of physiotherapy in reducingrespiratory complications following abdominal surgery. Physiotherapy is beneficialboth prophylactically, i.e. pre-operatively and also post-operatively. One largestudy concluded that low risk patients benefit from breathing excercises and highrisk patients benefit from incentive spirometry excercises.24

STRESS RESPONSE TO ABDOMINAL SURGERY

• The intensity of the stress response is related to the degree of tissuetrauma, i.e. minor surgery stimulates a minor, transient response whereasmajor abdominal surgery may stimulate a stress response lasting days toeven weeks. Other factors promoting the stress response after majorabdominal surgery include gut stimuli via the sympathetic nervous system, local tissue factors and cytokines. Haemorrhage, hypothermia,sepsis and acidosis will all exacerbate the response.

• The response is multifactorial, thus neural blockade will not completelyprevent this.

• The role of the stress response is to mobilise substrate and acute proteinsfor wound healing and the inflammatory response. Possible detrimentaleffects of a profound stress response following major surgery includeincreased demands on organs which may have reduced reserve, pul-monary complications, thromboembolism and pain and fatigue. Theappropriateness of an unmodified response is, therefore, debatable.

• Intraoperative regional anaesthesia may only delay the development ofthe stress response. The optimum duration of blockade is not known.

• If the response is desired to be modified into the post-operative perioda continuous regional technique is required – continuous epidural analgesia has been most studied.

• Epidural analgesia has significant modifying effects on the hormonaland catecholamine responses to lower abdominal surgery.

• The effects of epidural anaesthesia on the stress response following upperabdominal and thoracic surgery is less impressive. This could be due tofailure to adequately block all afferent stimulation. For example, continu-ous spinal anaesthesia,with its denser block, is more effective at blockingthe hormonal stress response compared with epidural anaesthesia.

• Central neural block may mitigate various aspects of post-operativemorbidity but the evidence is really only convincing for decreasedblood loss, reduction in DVT and limiting gastrointestinal stasis afterabdominal procedures (see also chapter on Regional Anaesthesia).


174

Chap-12.qxd 2/2/02 1:05 PM Page 174

• Spinal opioids have less effect on the stress response. Their effect onmorbidity is unclear but is likely to be less due to lesser effects on stressresponse.

SURGICAL ASPECTS

Inflammatory bowel diseaseMany of these patients will be very ill, pyrexial, dehydrated and septic. Nutritionalstate and wound healing will be poor. They may be young, have undergoneabdominal surgery before and be undergoing complex, prolonged reconstructivesurgery. If fistulae are present fluid, electrolyte and protein losses can be consider-able. The patients have often been managed pre-operatively with steroids andimmunosuppressive agents.

Perforated intra-abdominal viscusOften elderly with coexisting medical conditions. Many will be perforations ofdiverticular disease or malignancies. As the presentation may be unclear many languish for several days on medical wards before presenting to the surgeons withmarked sepsis. Large volumes of fluid, pus and/or faecal matter may be present inthe abdominal cavity. Operative mortality is of the order of 50%.

Bowel obstructionLarge volumes of fluid may be sequestered in the dilated loops of bowel. Withhigh obstruction the risk of aspiration at the induction of anaesthesia is marked.With prolonged obstruction perforation will occur leading to worse sepsis.Splanchnic blood flow will be reduced and inflammatory mediators released.

Percutaneous drainage of intra-abdominal abscessesRadiological techniques for drainage of intra-abdominal collections are constantlyadvancing. Unfortunately there are no prospective randomised trials comparing‘open’ drainage versus percutaneous drainage. Large retrospective comparisonssuggest that there are no differences in morbidity and mortality.25 Thus, it seemsappropriate to prefer radiologically guided percutaneous drainage of abscesses andother collections where possible.

Empirical laparotomy in post-operative sepsisIn the face of a patient deteriorating from sepsis on ICU who has had previousgastrointestinal surgery, surgeons may come under pressure to perform an empir-ical or ‘blind’ laparotomy to rule out intra-abdominal collection. However, there isno convincing evidence to support such an approach.26 Advances in radiologicaldiagnosis and radiologically guided drainage of collections have reduced the placefor such exploratory surgery.


175

Chap-12.qxd 2/2/02 1:05 PM Page 175

Resection of perforated colon versus drainage and colostomy formationAlthough technically a more demanding operation, resection of perforated colonseems preferable to drainage and colostomy formation. Several series have con-firmed this to be so. Pooled data suggest lower mortality, morbidity, hospital stayand rate of fistula formation.26

Primary anastomosis following colonic perforationAn anastomosis at the time of initial surgery prevents ongoing morbidity from acolostomy and the need for further surgery for closure. However, standard teach-ing has been that anastomotic breakdown is high in the face of an inflammatoryperitonitis. A recent review of pooled data from retrospective studies provide evi-dence that primary anastomosis is not associated with increased complications andmay even be preferable.26

Further reading

Richardson J, Sabanathan S. Prevention of respiratory complications after abdom-inal surgery. Thorax 1997; 52 (suppl. 3): S35–40.

Steinbrook RA. Epidural anesthesia and gastrointestinal motility. Anesth Analg1998; 86: 837–44.

Ogilvy AJ, Smith G. The gastrointestinal tract after anaesthesia. Eur J AnaesthesiolSuppl 1995; 10: 35–42.

References

1. Burnand KG, Young AE (eds), The New Aird’s Companion to Surgical Studies.London: Churchill Livingstone, 1998.

2. Krogh B, Jorn Jensen P,Henneberg SW et al. Nitrous oxide does not influenceoperating conditions or postoperative course in colonic surgery. Br J Anaesth1994; 72: 55–7.

3. Morimoto H, Cullen JJ, Messick JM et al. Epidural analgesia shortens post-operative ileus after ileal pouch-anal canal anastomosis. Am J Surg 1995; 169:79–82.

4. Vedrinne C, Vedrinne JM, Guirard M, Patricot MC, Bouletreau P. Nitrogensparing effect of epidural administration of local anaesthetics in colon surgery.Anesth Analg 1989; 60: 354–9.

5. Cali RL, Meade PG, Swanson MS et al. Effect of morphine and incisionlength on bowel function after colectomy. Dis Colon Rectum 2000; 43: 163–8.

6. Holte K, Kehlet H. Epidural analgesia and risk of anastomotic leakage. RegAnesth Pain Med 2001; 26: 111–17.


176

Chap-12.qxd 2/2/02 1:05 PM Page 176

7. Ryan P, Schweitzer S, Collopy B et al. Combined epidural and general anes-thesia versus general anesthesia in patients having colon and rectal anasto-moses. Acta Chir Scand Suppl 1989; 550: 146–9.

8. Morisot P, Loygue J, Guilmet C. Effects of postoperative decurarization withneostigmine on digestive anastomoses. Can Anaesth Soc J 1975; 22: 144–8.

9. Golub R, Golub RW, Cantu R et al. A multivariate analysis of factors con-tributing to leakage of intestinal anastomoses. J Am Coll Surg 1997; 184:364–72.

10. Mythen MG, Webb AR. Perioperative plasma volume expansion reduces theincidence of gut mucosal hypoperfusion during cardiac surgery. Arch Surg1995; 130: 423–9.

11. Houbiers JG, Brand A, van de Watering LM et al. Randomised controlled trialcomparing transfusion of leucocyte-depleted or buffy-coat-depleted blood insurgery for colorectal cancer. Lancet 1994; 344: 573–8.

12. The Veterans Affairs Total Parenteral Nutrition Cooperative Study Group.Perioperative total parenteral nutrition in surgical patients. N Engl J Med 1991;325: 525–32.

13. Buzby GP. Overview of randomized clinical trials of total parenteral nutritionfor malnourished surgical patients. World J Surg 1993; 17: 173–7.

14. Moore FA, Feliciano DV, Andrassy RJ et al. Early enteral feeding, comparedwith parenteral reduces postoperative septic complications. The results of ameta-analysis. Ann Surg 1992; 216: 172–83.

15. Morlion BJ, Stehle P, Wachtler P et al. Total parenteral nutrition with gluta-mine dipeptide after major abdominal surgery: a randomized, double-blind,controlled study. Ann Surg 1998; 227: 302–8.

16. Jones C, Palmer TE, Griffiths RD. Randomized clinical outcome study of critically ill patients given glutamine-supplemented enteral nutrition. Nutrition1999; 15: 108–15.

17. Heys SD, Walker LG, Smith I et al. Enteral nutritional supplementation withkey nutrients in patients with critical illness and cancer: a meta-analysis of randomized controlled clinical trials. Ann Surg 1999; 229: 467–77.

18. Ballantyne JC, Carr DB, de Ferranti S et al. The comparative effects of post-operative analgesic therapies on pulmonary outcome: cumulative meta-analysesof randomized, controlled trials. Anesth Analg 1998; 86: 598–612.

19. Tweed WA,Phua WT,Chong KY et al. Large tidal volume ventilation improvespulmonary gas exchange during lower abdominal surgery in Trendelenburg’sposition. Can J Anaesth 1991; 38: 989–95.


177

Chap-12.qxd 2/2/02 1:05 PM Page 177

20. Pelosi P,Ravagnan I,Giurati G et al. Positive end-expiratory pressure improvesrespiratory function in obese but not in normal subjects during anesthesia andparalysis. Anesthesiology 1999; 91: 1221–31.

21. Pedersen T, Viby-Mogensen J, Ringsted C. Anaesthetic practice and post-operative pulmonary complications. Acta Anaesthesiol Scand 1992; 36: 812–18.

22. Tulla H, Takala J,Alhava E et al. Breathing pattern and gas exchange in emer-gency and elective abdominal surgical patients. Inten Care Med 1995; 21:319–25.

23. Lacy PD, Burke PE, O’Regan M et al. The comparison of type of incision fortransperitoneal abdominal aortic surgery based on postoperative respiratorycomplications and morbidity. Eur J Vasc Surg 1994; 8: 52–5.

24. Hall JC,Tarala RA,Tapper J et al. Prevention of respiratory complications afterabdominal surgery: a randomised clinical trial. Br Med J 1996; 312: 148–52.

25. Bufalari A,Giustozzi G,Moggi L. Postoperative intraabdominal abscesses: per-cutaneous versus surgical treatment. Acta Chir Belg 1996; 96: 197–200.

26. Jimenez MF, Marshall JC. Source control in the management of sepsis. IntenCare Med 2001; 27: S49–62.


178

Chap-12.qxd 2/2/02 1:05 PM Page 178

13PERIOPERATIVE RENAL

INSUFFICIENCY AND FAILURE

Surgery, especially major surgery, performed on high risk or very ill patients iscommonly associated with a worsening of renal function. Indeed, the develop-ment of renal failure in the postoperative period is a major source of mortality.The purpose of this chapter is to review causes, pathogenesis, prevention and man-agement of renal insufficiency and failure in surgical patients. Some of the princi-ples and controversies of artificial renal support in the intensive care unit arebriefly discussed.

DEFINITIONS

Renal insufficiency may be thought of as a worsening of renal function either interms of reduced urine flow or an increase in urea and creatinine compared tobaseline or normal values. Renal insufficiency in the postoperative period is com-mon following major surgery. Elderly patients may also have decreased renal func-tion and can be thought of as having renal insufficiency or reduced renal reserve.Renal insufficiency is a useful concept as it identifies patients who may be atincreased risk of progressive worsening of renal function leading to renal failure.

Acute renal failure (ARF) has received many definitions in the medical literature.Unfortunately this makes comparisons of research trials difficult and meta-analysisimpossible! One reason for this is that there is a continuum of renal function withprogressively falling creatinine clearance.

Thus patients may present at any point from ‘normal’ to complete anuria/nointrinsic renal function. Different definitions focus on different points in this con-tinuum:

Normal Progressive Anuric/no renal insufficiency intrinsic function

179

Chap-13.qxd 2/1/02 12:09 PM Page 179

Useful clinical definitions of ARF include:

• Oliguria with urine less than 0.5 ml/kg/h. In severe oliguria the urineflow will be less than 400 ml/day.

• Rapidly rising urea or creatinine. Doubling of serum creatinine is significant in some definitions.

• Creatinine clearance less than 20 ml/min.

• Lesser degrees of oliguria or elevations of creatinine with a metabolicacidosis (excluding diabetic ketoacidosis or lactic acidosis).

ARF is often subdivided into three categories:

• Pre-renal – including true hypovolaemia following, e.g. haemorrhagebut also other causes of reductions in renal blood flow (RBF), e.g. cardiacfailure. Certain drugs causing vasoconstriction in the afferent arteriolesalso come into this category, e.g. ACE inhibitors and nonsteroidal anti-inflammatory drugs (NSAIDs).

• Renal – including all the ‘medical’ causes of ARF which are beyondthe scope of this text but also includes tubular damage. Tubular damage(acute tubular necrosis) will follow from persistent reductions in RBFas above but also from tubular toxins such as aminoglycoside antibiotics,myoglobin and X-ray contrast medium.

• Post-renal – prolonged obstruction to any part of the urinary systemwill lead to ARF. Causes include enlarged prostate, stones and tumours.Prompt relief of the obstruction is essential.

Two points are worth emphasising:

1. Pre-renal causes predominate in surgical patients.

2. All patients with ARF must have post renal obstruction excluded, e.g. byabdominal ultrasound.

Urinary electrolytes

Urinary electrolytes and osmolality have been used traditionally to help distin-guish pre-renal from renal ARF:

However, this is no more than a guide and cannot always be relied upon. The useof diuretics which increase Na excretion abolish the value of this investigation.


180

Urinary Na Urinary osmolality

Pre-renal � 20 � 500Renal � 20 � 350

Chap-13.qxd 2/1/02 12:09 PM Page 180

Difficulties in measurements and diagnosis

• Creatinine clearance is an inexact measure of glomerular filtration rate (GFR). Problems arise when collecting urine over 24 h especiallyin non-catheterised patients and, of course the result is received morethan 24 h after decision is taken to perform the investigation! A collec-tion over 2 h is probably as accurate especially in catheterised patients inwhom one can be sure there is no volume missed because of residualurine in the bladder.

• Creatinine clearance can be estimated using the Cockroft formula fromthe serum creatinine. This is not reliable in the postoperative or critic-ally ill patient in whom creatinine may be rising rapidly or large fluidshifts are occurring.

Potential pitfalls in diagnosis

• The kidneys have tremendous reserve. GFR can fall by 70% before creatinine will start to rise. Thus any elevation of creatinine representsrenal insufficiency.

• Creatinine will be lower and rise slower in patients with reduced muscle mass.

• Urea is also used as a guide to renal function. However, will beincreased in dehydration (in itself a valuable sign), following GI bleedand with high protein intake. Probably more useful as a guide to thetiming of dialysis in order to avoid uraemic complications.

Despite these reservations creatinine and creatinine clearance remain the mostuseful measures of renal function in common clinical usage.

RENAL FAILURE IN SURGICAL AND CRITICALLY ILLPATIENTS – A SEPARATE DISEASE?

Kishen1 makes a persuasive argument that ARF in the surgical or ICU setting is adifferent disease from ‘medical’ renal failure found in the nephrology wards. Littleis known for certain about the underlying pathology as few critically ill patientsever have a renal biopsy undertaken for histological analysis. Many such patients arestated to have acute tubular necrosis but this is rarely substantiated histologically.

ARF in the critically ill is usually part of a multiple organ dysfunction syndromeand rarely an isolated single organ disease. It is often associated with sepsis, is multi-factorial in aetiology (e.g. a combination of sepsis and hypovolaemia) and has ahigh mortality. On the other hand ‘medical’ ARF usually is a single organ disease,has a low mortality and often is the result of a distinct insult (e.g. autoimmune disease, drug toxicity, etc.). Surviving patients recovering from ARF in ICU usually

PERIOPERATIVE RENAL INSUFFICIENCY AND FAILURE

181

Chap-13.qxd 2/1/02 12:09 PM Page 181

do not require long term renal support whereas in ‘medical’ ARF many patientsprogress to chronic renal failure (CRF).

Incidence

The incidence of ARF in hospital patients is unclear owing to differences in defin-itions in studies performed.

• Community acquired ARF comprises only approximately 1% of alladmissions to hospital.

• In one large study the incidence of hospital acquired ARF was approxi-mately 5%. However, other studies suggest that in subgroups such assurgical and radiological patients receiving X-ray contrast media theactual incidence of ‘iatrogenic’ ARF may be as high as 50%.2

• Following cardiac arrest, 30% will develop ARF. The incidence riseswith increasing duration of resuscitation and increasing doses of adrenaline administered.

• The incidence of ARF in general ICUs varies between 10% and 25%depending on the local patient populations. The majority of these arepatients with surgical pathology. ‘Medical’ ARF is much less common.Associated risk factors include sepsis, hypotension and iatrogenic factorssuch as the administration of X-ray contrast and aminoglycosides. Inone study hypotension was a factor in 85% of patients developing ARFon the ICU and was the sole factor in 33%.3

The majority of cases of ARF now occur in ICU leading some to question therole of nephrologists in the management of ARF in the ICU. However, the role ofnephrologists in diagnosis and therapy of ‘medical’ ARF remains unquestioned.

The only systematic review into preoperative risk factors bemoaned the lack ofconsistency in definitions and statistical analysis and could make few firm conclu-sions.4 Repeatedly studies have identified preexisting renal dysfunction, poor LVfunction and advanced age as the main predictors of postoperative ARF in generalsurgical populations.

The incidence of ARF is much higher in certain surgical settings and these con-stitute the high risk groups for development of ARF in the perioperative period:

Cardiac surgeryFifteen per cent of cardiac surgery patients may experience elevations of creatininein the postoperative period. There seems to be a linear relationship betweenbypass time and ARF.

With a more stringent definition of ARF a recent study of over 42 500 patientsfound an overall incidence of ARF of 1.1%.5 It is important to note that ARF


182

Chap-13.qxd 2/1/02 12:09 PM Page 182

following cardiac surgery is a major determinant of mortality – 63.7% mortality in those patients with ARF compared to only 4.3% in whom renal functionremained normal. The conclusion of this study is that ARF is independently asso-ciated with early mortality after cardiac surgery even after adjustments for comor-bidity and postoperative complications.

Vascular surgerySeveral studies have identified that vascular surgery and especially aortic surgery is a major risk factor for perioperative renal dysfunction. Risk factors within thisgroup are consistent between the various studies and include:

• advanced age,

• elevated creatinine preoperatively,

• large volume of transfused blood (itself equating with length of surgeryand difficulty),

• duration of aortic cross clamping,

• requirement for postoperative ventilation and/or inotropes.

It would seem that the approximate incidence of renal impairment after aortic sur-gery is in the order of 25% with a high mortality.

Surgery in liver diseaseARF occurs in approximately 10% of patients operated on with obstructive jaundice.

Almost 70% of patients with severe liver failure, e.g. undergoing liver transplant-ation develop ARF.

OTHER RISK FACTORS

• Elderly patients also constitute a high risk group owing to their reducedcardiorespiratory and renal reserve although figures on incidence areless certain.

• Rhabdomyolysis: Myoglobin from muscle breakdown precipitates out ofsolution and blocks the renal tubules.

• NSAIDs: Block the vasodilator effect of prostaglanding. A recent paperlooking at the epidemiology of ARF found that NSAIDs were used in18% compared with 11% of control patients, i.e. the link is probablyimportant.6

• Abdominal compartment syndrome: The presence of increased intraabdomi-nal pressure from ileus, haematoma, intraabdominal packs literally


183

Chap-13.qxd 2/1/02 12:09 PM Page 183

squeezes the kidney. This causes a reduction in RBF, GFR, direct com-pression of the renal parenchyma and increased release of ADH and aldos-terone from stimulation of abdominal wall stretch receptors. In generalintraabdominal pressures of 15–20mmHg are associated with oliguriawhile pressures greater than 30mmHg may be associated with anuria.

• X-ray contrast medium: The risk of renal dysfunction is worse if thepatient is diabetic or has baseline renal dysfunction.

• Antibiotics especially aminoglycosides: It seems that the rate of renal corticaluptake of gentamicin is saturatable, i.e. lesser side effects including renaldysfunction if the total daily dose is given once rather than in divideddoses.

• Excess use of diuretics: May contribute to hypovolaemia.

• ACE inhibitors: ACE inhibitors induce ARF in renal artery stenosis.Function often improves when the ACE inhibitors are stopped. Therise in creatinine is worse if the patient is also taking diuretics.

• Aprotinin: There were reports of renal problems from high doses of theprotease inhibitor, aprotinin, in cardiac surgery patients. However, themost recent study found no effect of aprotinin on renal function duringhypothermic cardiopulmonary bypass.

• ‘Medical’ risk factors: The incidence of renal dysfunction and its severitywill be increased in the presence of recognised risk factors for renal dys-function such as hypertension and diabetes mellitus. The presence ofintrinsic renal disease or chronic renal failure will also obviously increasethe likelihood of perioperative deterioration in renal function.

Problem with colloids?

The choice of crystalloids versus colloids for fluid resuscitation has been hotlydebated for over 20 years with most reasonable practitioners accepting a role forboth colloids and crystalloids depending on circumstances. Even the colloid prot-agonists accept that there is no firm evidence that use of colloid infusions improvesmeasures of outcome. In recent years concerns have been raised that the overuseof colloid fluids may be associated with a worsening of renal function. This hasbeen termed ‘hyperoncotic acute renal failure’. The original reports concernedthe use of dextrans (and at that time it was thought that there was either direct tox-icity or accumulation in the renal tubules) but may be seen with excess use of allcolloids especially if the patient

• has other renal risk factors,

• is dehydrated, and


184

Chap-13.qxd 2/1/02 12:09 PM Page 184

• is worse with use of colloids in large volumes or high molecular weights.

The theory proposed is that the resultant high plasma colloid osmotic pressurecounteracts opposing hydrostatic filtration pressure in the glomerulus.

Interestingly, small studies suggest that colloids should not be used as plasma volume expanders in brain dead organ donors or kidney transplant recipients. Thefunction of the kidney may be at risk.

Factors involved in pathogenesis

• In high risk surgical and critically ill patients the main factors involvedin pathogenesis are reductions in cardiac output, RBF and GFR.

• Hypovolaemia and haemorrhage are obviously potent mechanisms forinterfering with the functioning of the kidney.

• Because cardiac output is such an important factor in ARF, cardiac riskfactors also predict for perioperative renal insufficiency and develop-ment of ARF.

• Inflammatory mediators and cytokines, especially in septic patients, alsoreduce RBF and GFR partly through interference with prostaglandinpathways. Thus, development of surgical sepsis is an important risk fac-tor for perioperative renal insufficiency.

Under normal circumstances a single, short insult does not result in ARF inpatients with previously normal renal function. However, repeated and prolongedinsults will produce renal dysfunction and if corrective measures are not taken thepatient is more likely to develop ARF. A single insult may be enough to produceARF in a ‘population at risk’.

Influence of anaesthetic technique

Choice of anaesthetic technique itself is not thought to be a significant factor.Historically, volatile agents which released inorganic fluoride were a problem butthis is not an issue with modern halogenated agents such as isoflurane and sevoflu-rane. Techniques which preserve RBF are preferable. Central neural blockade,e.g. epidural may have a beneficial effect on RBF due to the reduction in periph-eral resistance provided hypotension and hypovolaemia are avoided. Conversely,cardiac output and RBF may fall with general anaesthesia. There will be muchinterpatient variation.

WHY THE KIDNEY IS AT RISK?

Renal problems are common in high risk surgical and critically ill patients.In addition, ARF is an early manifestation of developing multiple organ failure.


185

Chap-13.qxd 2/1/02 12:09 PM Page 185

This is despite the kidney arguably being the best perfused and oxygenated organin the body – receiving almost 25% of the total cardiac output. This high bloodflow is necessary to provide the high volumes of filtrate required for removal ofwaste products. Despite this high blood flow the kidney is usually the first organto fail in shock or multiple organ failure. Why is this? Why is the kidney so vul-nerable that it has been referred to as the body’s ‘innocent bystander’?

Table 13.1 illustrates the problem and its mechanisms.

Thus although the kidney itself has a relatively low O2 consumption to deliveryratio, the renal medulla (the metabolically active part of the kidney) has a very highratio. Indeed, the highest ratio in the body. Higher than the heart, an organ anaes-thetists are often preoccupied with especially regarding the possible developmentof ischaemia. Unfortunately, there are no symptoms or immediate signs of renalmedullary ischaemia.

Kidneys are also perfusion pressure dependent. Critically ill patients especiallywith sepsis lose their ability to autoregulate their RBF and this becomes linearwith blood pressure.

Renal defence mechanisms

Tubuloglomerular feedback (TGF) in the juxtaglomerular apparatus in the nephronis a protective mechanism which attempts to protect the kidney when perfusion islow. Efferent glomerular constriction reduces GFR to reduce the oxygen demandon the medulla. This also preserves vascular volume in hypovolaemia. The con-troversial role of increasing RBF versus improving perfusion pressure is discussedbelow.

WHAT MAY PREVENT RENAL FAILURE?

Good perioperative care and attention to detail will be the best chance of avoid-ing deterioration in renal function and development of renal failure.


186

Table 13.1 – Oxygen balance in different organs (measurements in ml/min/100 g).

Organ O2 delivery Blood flow O2 consumption O2 consumption/delivery ratio

Kidney 84.0 420 6.8 8Kidney medulla 7.6 190 6.0 79*Hepatoportal 11.6 58 2.2 18Brain 10.8 54 37 34Muscle 0.5 2.7 0.18 34Heart 16.8 84 11.0 65

Reproduced with permission from Brezis M, Rosen S, Epstein FH. Acute renal failure. In Brenner BM, Rector FC Jr (eds), The Kidney, 4th edn. Philadelphia: WB Saunders, 1991: 993–1061.

Chap-13.qxd 2/1/02 12:09 PM Page 186

Fluid therapy

The role of adequate fluid therapy and maintenance of blood volume as the cor-nerstone has been re-emphasised in a recent editorial.7 However, proof of efficacyis often lacking. However, in certain specific situations evidence for the effective-ness of intravenous fluids has been forthcoming.

• Sodium bicarbonate in rhabdomyolysis prevents myoglobin precipita-tion in the renal tubules by maintaining an alkaline urine.

• X-ray contrast induced renal dysfunction. There are several randomisedtrials showing that IV hydration reduces the incidence of renal dys-function associated with the administration of X-ray contrast media.Such studies are relatively easy compared to other groups of patientswith renal dysfunction as the insult and its timing are relatively pred-ictable and controllable. Although the results may not be assumed to apply to other models of renal dysfunction they are of interest. Manyof these and other studies have involved crystalloid infusions usually of saline. The role of colloid infusions is questionable as discussed previously.

Oxygen transport

In high risk surgical patients the application of an aggressive strategy of maintain-ing cardiac output, oxygen delivery and oxygen consumption may improve outcome. One of the original randomised controlled trials (RCTs) showed thatthe improvement in survival was associated with a reduction in organ failures, inparticular, renal failure.8 This is not surprising when one considers the reasons discussed above as to why the kidney is at especial risk compared to other organs.

Short cross clamp time in aortic surgery

During aortic cross clamping RBF is reduced even when the clamp is appliedbelow the origin of the renal arteries. Thus it is important to minimise the dur-ation of aortic clamping. Once the clamp is released the reperfusion of the lowerlimbs will release mediators and oxygen radicals which may impair renal function.

Short bypass time in cardiac surgery

Similarly, as short a time as possible spent on cardiopulmonary bypass is importantas RBF is reduced during bypass and red blood cells are damaged. Cellular debrisand free haemoglobin are damaging to the kidney.

Frusemide

Animal studies clearly indicate that frusemide is protective to the kidney.Patient studies do not. Many believe that conversion of oliguric renal failure to


187

Chap-13.qxd 2/1/02 12:09 PM Page 187

non-oliguric renal failure by use of diuretics is beneficial. As summarised above inTable 13.2 the mortality rate for oliguric versus non-oliguric renal failure seemsbetter. This does not necessarily mean that one can improve an individual patient’soutcome with diuretics as there is evidence that those patients who respond todiuretics probably have less severe renal damage. Among those patients whorequire renal support mortality is not improved by the use of diuretics.

The use of infusions of frusemide rather than intermittent boluses may be prefer-able because of decreased requirement for equivalent effect, improved responseand less tolerance with few adverse effects.

Mannitol

The osmotic diuretic mannitol is widely used to promote urine flow. Suggestedmechanisms include its osmotic action with greater washout of solutes, plasmavolume expansion and increased RBF. Animal studies suggest improved renalfunction but patient studies especially in the high risk surgical patient are less con-vincing. Many studies demonstrate increased urine volumes but are less convin-cing in terms of preservation of renal function.

Mannitol (in conjunction with bicarbonate) seems beneficial in the special situ-ation of rhabdomyolysis. However, IV hydration is superior to hydration pluseither mannitol or frusemide in X-ray contrast induced renal dysfunction.9

Dopexamine

Dopexamine is a relatively new addition to the debate on perioperative renal protection. Its actions on dopaminergic receptors undoubtedly increase RBF and creatinine clearance in animal studies and volunteers but its role and efficacyin prevention of renal failure in critically ill patients is less established. Studies in cardiac surgery and vascular surgery patients indicate some protective


188

Table 13.2 – Mortality in non-oliguric versus oliguric renal failure.

Number of patients in study* Non-oliguric mortality (%) Oliguric mortality

143 28 78462 58 65151 42 83109 17 52228 42 65125 23 63

* � 450 ml urine/24 h defined as non-oliguric. These studies were in general ARF populations.Reproduced with permission from Thadhani RI, Bonventre JV. Acute renal failure. In Lee BW, Hsu SI,Stasior DS (eds), Quick Consult Manual of Evidence Based Medicine. Philadelphia: Lippincott-Raven,1997: 382–413.

Chap-13.qxd 2/1/02 12:09 PM Page 188

effect. Dopexamine may have other beneficial effects in the high risk surgicalpatient:

• Reduces the usual increase in gut permeability after cardiac surgery.

• May have specific anti-inflammatory actions in surgical patients.

Charing Cross protocol

A protocol developed at Charing Cross hospital in the 1990s has been claimed todramatically reduce the need for haemofiltration in ARF in ICU patients. Thereare three main aspects of this protocol which, in truth, just restate many of theprinciples already described:

1. Achieving normovolaemia with monitoring guided by pulmonary arterycatheter and echocardiographic measures of preload. GTN infusions arealso used to encourage vasodilation and to reduce myocardial ischaemia.

2. Achieving normotension (for the patient’s age) with emphasis on the roleof noradrenaline.

3. Reducing oxygen consumption in the renal medulla with low dose infu-sions of frusemide.

Although theoretically attractive, widely followed and (anecdotally) often veryeffective this protocol has never been validated by large RCTs.

RBF VERSUS PERFUSION PRESSURE

• Many of our therapies aimed at improving urine production and renalfunction do so, we presume, by improving RBF. Unfortunately therelationship between RBF and GFR is complex, varies from patient topatient and is poorly understood. Thus dopamine, dopexamine anddobutamine, all widely used for this purpose, may have inconsistenteffects clinically. The optimisation of cardiac output and oxygen trans-port as mentioned above presumably acts in this context by increasingRBF.

• The ‘downside’ of increasing RBF is that this presents more solute forclearance with the requirement for increased Na reabsorption in themedulla and increased medullary oxygen consumption. It has even beensuggested that the oliguria resulting from reductions in RBF protectsthe kidney by reducing medullary oxygen consumption!

• Conversely, increasing blood pressure improves renal perfusion pres-sure and may improve GFR.10 Thus increasing blood pressure withnoradrenaline (norepinephrine) may improve creatinine clearance.


189

Chap-13.qxd 2/1/02 12:09 PM Page 189

For example, there are numerous studies of septic shock demonstratingincreased urine flow and creatinine clearance in patients treated withthe addition of noradrenaline to standard therapies. The caveats are thatcardiac output and RBF must be maintained (necessitating monitoringof cardiac output) and hypovolaemia prevented. The ‘Charing Cross’protocol involves as one of its key elements the judicious use of noradrenaline.

The above may seem confusing but unfortunately many of the controversies andapparent contradictions remain unresolved.

WHAT MAY NOT PREVENT RENAL FAILURE?

Diuretics – the dual edged sword

The maintenance of urine flow does seem to exert some preservative effect onrenal function – at least in certain subgroups such as patients at high risk of X-raycontrast medium induced renal failure. However, aggressive hydration of thesepatients has been shown to exert a better protective effect than hydration plusdiuretics.9 In general patients do not go into ARF because they are lacking infrusemide! The diuresis produced may be harmful by giving a false sense of secu-rity and by exacerbating any preexisting hypovolaemia.

Dopamine – its hard to keep this drug down

Despite more than 20 years of experience and clinical trials the use of ‘renal dosedopamine’ is still dogged with controversy. Dopamine is still widely used, at lowrates of infusion, as an agent to protect the kidney and prevent renal failure. Thisis despite numerous editorials and review articles stating that there is no renal protective effect of dopamine.

Dopamine has been claimed to exhibit different effects depending on the infusionrate:

• 0.5–3 �g/kg/min – dopaminergic effect increasing RBF

• � 3 �g/kg/min – �1 effect increasing cardiac output

• � 7 �g/kg/min – increasing �1 effect leading to vasoconstriction.

This itself represents a realisation that vasoconstriction occurs at lesser doses thanwas originally believed, i.e. older texts refer to vasoconstriction occurring at� 15 �g/kg/min.

The so-called renal dose and effects of increasing dosage described in the textbooks have always been suspected to be less predictable in the real world. Sureenough it has now been demonstrated in volunteers that plasma levels ofdopamine vary widely for identical infusion rates in terms of �g/kg/min.11


190

Chap-13.qxd 2/1/02 12:09 PM Page 190

Critically ill patients would be presumed to be even more varied in their drug dis-position and metabolism. Thus there is now no real justification to think of onedose of dopamine as being a ‘renal dose’:

• In volunteers dopamine infusions lead to increased RBF, urine produc-tion and creatinine clearance. Animal studies suggest a beneficial effecton renal function – renal protection (the holy grail of nephrology).

• There is still no convincing evidence to support the idea that the use of dopamine at low dosage prevents the development of ARF inpatients.

• What few trials there are which seem to show benefit, especially forradiological contrast medium induced ARF include vigorous fluidloading as part of their protocol.

• Perioperative use of dopamine has been widespread, e.g. cardiac andaortic surgery but this is not supported by scientific evidence.

• Dopamine does increase urine output, mainly by a natriuretic ordiuretic effect but also by a general increase in cardiac output (Howeverdobutamine does this better due to its more predictable inotropic properties).

• A recent, large, well conducted study failed to demonstrate any benefitof low dose dopamine in critically ill patients at risk of renal failure.12

Many have in the past used renal dose dopamine for two reasons:

1. It may be renal protective, they just haven’t been able to prove it.

2. It may not do any good but at least it does not do any harm.

Never mind this latter unusual justification for giving any drug, there is an increas-ing opinion that dopamine is potentially harmful due to

• tachycardia and arrhythmias,

• reversible, dose related depression of pituitary release of prolactin anddepression of indices of cell mediated immunity,13

• the diuresis may be harmful by giving false sense of security and byexacerbating hypovolaemia,

• a recent randomised study on the use of dopamine for X-ray contrastmedium induced renal failure demonstrated a harmful effect on theseverity of renal failure, prolonging the course,14

• dopamine seems to have worse effects on the splanchnic circulationcompared to a pure vasoconstrictor, noradrenaline.


191

Chap-13.qxd 2/1/02 12:09 PM Page 191

EXPERIMENTAL DRUGS TO PREVENT RENAL FAILURE

Many agents have been studied in animal models as either prevention or treatmentof ARF. Less of these agents have made it to clinical trials and none have convin-cingly been shown to be of clinical benefit:

• Atrial natriuretic peptide (ANP), endothelin antagonists and calciumchannel blockers have shown most promise but further studies are needed.

• Concerns exist with some of these agents, e.g. calcium channel blockersthat interaction with anasesthetics agents may produce more hypoten-sion which may mitigate against any renal protective effect.

• Other therapies suggested by animal trials, e.g. thyroxine have not beensubstantiated in patient studies.

• Hypotensive complications of these agents may limit their usefulness.

• Studies have on occasion been poorly controlled with, e.g. frequentuncontrolled concomitant use of diuretics and dopamine.

• At present all these experimental drugs remain just that, experimental.

ESTABLISHED ARF

Once ARF has occurred there are no known pharmacological interventionswhich will reverse the process – including diuretics. The kidney will eitherrecover or it will not (though interesting new evidence suggests that the methodof supportive therapy may influence this as discussed below). Thus the mainstay ofmanagement is supportive measures both in the acute stages and later if the ARFbecomes ‘chronic’.

A full discussion of renal replacement techniques is beyond the scope of this textbut some interesting results from recent studies will be highlighted especially asthey offer some advice as to whether there are any differences in outcome betweendifferent techniques of renal support.

Indications for renal support include:

• Uraemia, a blood urea � 50 is an arbitrary but common guide torequirement for support. Uraemic complications may occur withhigher blood urea including pericarditis and GI bleeding. Support shouldperhaps be instituted earlier in malnourished patients with reducedmuscle mass whose urea and creatinine will rise slower.

• Severe acidosis.

• Hyperkalaemia.

• Pulmonary oedema secondary to fluid overload in oliguric patient.


192

Chap-13.qxd 2/1/02 12:09 PM Page 192

In modern critical care practice any of the above should be sufficient to triggerrenal support unless there are definite contraindications to its provision.

Recent studies examining the role of early versus late intervention in terms ofproviding renal support are of interest. For example:

• Early initiation of continuous venovenous haemofiltration (CVVH) ina small study of patients with septic shock has been shown to improvehaemodynamic and metabolic responses and improve survival.15

• Early haemofiltration is associated with better than predicted survival inARF after cardiac surgery.16

• Despite similarities in injury severity and risk of developing renal fail-ure, survival was increased in the early filtration group in a retrospectivereview of 100 trauma patients.17

Thus, many ICUs institute support at an early stage before biochemical decom-pensation occurs.

Continuous versus intermittent replacement therapies

A majority of modern ICUs offer continuous renal support usually CVVH. Somestill offer only intermittent dialysis with some units offering both or neither.The advantages of CVVH include:

• ‘Gentler’. The fluid shifts are gradual and continuous leading to greaterhaemodynamic stability.

• Gradual removal of solutes avoiding the potential for disequilibriumsyndrome.

• Experimental role in removing inflammatory mediators in sepsis andpancreatitis.

Recent studies provide evidence that CVVH is better than intermittent dialysis:

• The rapid fluid shifts and haemodynamic upset associated with dialysisare associated with further ischaemic insults to the kidney in animals.

• Creatinine clearance falls after dialysis! (but does not after CVVH) –presumably because of these shifts.18

• CT studies of the brain demonstrate changes after dialysis consistentwith increased brain water which are not present after CVVH.19

• A RCT has shown that intrinsic renal function among survivors ofARF recovers to a predetermined level in over 90% of patients treatedwith CVVH but in only 59% of patients treated with dialysis (i.e. more


193

Chap-13.qxd 2/1/02 12:09 PM Page 193

patients will require chronic renal support when ARF is managed withintermittent dialysis).20

• Several studies support the observation that the duration of ARF is lesswhen the patient is managed with CVVH than with dialysis.

• Meta-analysis of (mainly) retrospective studies show improved survivalwith continuous techniques. Large RCTs are awaited.

High volume haemofiltration?

The volume of ultrafiltrate may be important. Large volumes of ultrafiltrate (high volume or ‘aggressive’ CVVH) increase creatinine clearance and mayremove greater quantities of inflammatory mediators. The only large RCT of over400 patients demonstrated improved survival with high volumes of ultrafiltratecompared to lower volumes.21 The authors recommend ultrafiltrate volumes of atleast 35 ml/kg/h.

OUTCOME OF ARF

• ‘Medical’ renal failure has a better outcome than ‘surgical’ ARF orARF in the critically ill patient. One study from 1997 found a 16%mortality in the ‘nephrotoxic’ group compared to a 30% mortality inthe ‘ischaemic’ group.22

• Many studies have found the mortality of ARF in the ICU to beapproximately 40–70%. When ARF is associated with the need forIPPV (i.e. part of the multiple organ failure syndrome) the mortality isof the order of 70%.

• Mortality is high when the onset of ARF is late in the course of a crit-ical illness, i.e. a recent French multicentre trial of over 1000 patientsfound that mortality was worse in patients developing ARF in ICUcompared to patients admitted to ICU with ARF.23

• There is cause for optimism, however, from the studies quoted abovecomparing the relatively newer technique of CVVH with dialysis. Inaddition comparisons of different cohorts treated for ARF at the MayoClinic have been published:24


194

Patients Survival (%)

1977–79 71 321991–92 71 52

Chap-13.qxd 2/1/02 12:09 PM Page 194

There was a poor prognosis if sepsis and the requirement for IPPV werecombined but overall this suggests an improving prognosis for ARF.

• In addition a recent study of over 300 patients from Holland managedwith high volume CVVH had a mortality of 33% compared with apredicted mortality of 67% given by the Madrid ARF score.25

However, it is still better to prevent ARF in the high risk surgical patient than totreat it.

Further reading

Bellomo R, Ronco C (eds), Acute Renal Failure in the Critically Ill. Berlin: SpringerVerlag, 1995.

Galley HF (ed.), Critical Care Focus 1: Renal Failure. BMJ Books, 1999.

References

1. Kishen R. Acute renal failure. In McConachie I (ed.), Handbook of ICUTherapy. London: Greenwich Medical Media, 1999: 161–72.

2. Hou SH, Bushinsky DA, Wish JB et al. Hospital acquired renal insufficiency:a prospective study. Am J Med 1983; 74: 243–8.

3. Menash PL, Ross SA, Gottlieb JE. Acquired renal failure in critically illpatients. Crit Care Med 1988; 16: 1106–9.

4. Novis BK, Roizen MF, Aronson S, Thisted RA. Association of preoperativerisk factors with postoperative acute renal failure. Anesth Analg 1994; 78:143–9.

5. Chertow GM,Levy EM,Hammermeister KE. Independent association betweenacute renal failure and mortality following cardiac surgery. Am J Med 1998;104: 343–8.

6. Griffin MR, Yared A, Ray WA. Nonsteroidal antiinflammatory drugs andacute renal failure in elderly persons. Am J Epidemiol 2000; 151: 488–96.

7. O’Leary MJ, Bihari DJ. Preventing renal failure in the critically ill. Br Med J2001; 322: 1437–9.

8. Shoemaker WC, Appel PL, Kram HB et al. Prospective trial of supranormalvalues of survivors as therapeutic goals in high-risk surgical patients. Chest1988; 94: 1176–86.

9. Solomon R,Werner C,Mann D et al. Effects of saline,mannitol, and furosemideto prevent acute decreases in renal function induced by radiocontrast agents.N Engl J Med 1994; 331: 1416–20.


195

Chap-13.qxd 2/1/02 12:09 PM Page 195

10. Hoogenberg K, Girbes ARJ. The use of dopamine and norepinephrine inICU patients with special reference to renal function. Care Crit Ill 2000; 16:174–9.

11. Macgregor DA, Smith TE, Prielipp RC et al. Pharmacokinetics of dopaminein healthy male subjects. Anesthesiology 2000; 92: 338–46.

12. Bellomo R, Chapman M, Finfer S et al. Low-dose dopamine in patients withearly renal dysfunction: a placebo controlled randomised trial. Lancet 2000;356: 2112–13.

13. Bailey AR, Burchett KR. Effect of low-dose dopamine on serum concentra-tions of prolactin in critically ill patients. Br J Anaesth 1997; 78: 97–9.

14. Abizaid AS, Clark CE, Mintz GS et al. Effects of dopamine and aminophyllineon contrast-induced acute renal failure after coronary angioplasty in patientswith preexisting renal insufficiency. Am J Cardiol 1999; 83: 260–3.

15. Honore PM, Jamez J, Wauthier M et al. Prospective evaluation of short-term,high-volume isovolemic hemofiltration on the hemodynamic course and out-come in patients with intractable circulatory failure resulting from septicshock. Crit Care Med 2000; 28: 3581–7.

16. Bent P, Tan HK, Bellomo R et al. Early and intensive continuous hemofiltra-tion for severe renal failure after cardiac surgery. Ann Thorac Surg 2001; 71:832–7.

17. Gettings LG, Reynolds HN, Scalea T. Outcome in post-traumatic acute renalfailure when continuous renal replacement therapy is applied early vs. late.Inten Care Med 1999; 25: 805–13.

18. Manns M,Sigler MH,Teehan BP. Intradialytic renal hemodynamics: potentialconsequences for the management of the patient with acute renal failure.Nephrol Dial Transplant 1997; 12: 870–2.

19. Ronco C, Bellomo R, Brendolan A et al. Brain density changes during renalreplacement in critically ill patients with acute renal failure. Continuoushemofiltration versus intermittent hemodialysis. J Nephrol 1999; 12: 173–8.

20. Mehta R, McDonald B, Gabbai F et al. Acute renal failure in the ICU: resultsfrom a randomised multicentre trial. Am J Soc Nephrol 1996; 5: 7.

21. Ronco C, Bellomo R, Homel P et al. Effects of different doses in continuousveno-venous haemofiltration on outcomes of acute renal failure: a prospectiverandomised trial. Lancet 2000; 356: 26–30.

22. Weisberg LS, Allgren RL, Genter FC et al. Cause of acute tubular necrosisaffects its prognosis. The Auriculin Anaritide Acute Renal Failure StudyGroup. Arch Intern Med 1997; 157: 1833–8.


196

Chap-13.qxd 2/1/02 12:09 PM Page 196

23. Guerin C, Girard R, Selli Jean-Marc et al. Initial versus delayed acute renalfailure in the Intensive Care Unit. Am J Respir Crit Care Med 2000; 161:872–9.

24. McCarthy JT. Prognosis of patients with acute renal failure in the intensive-care unit: a tale of two eras. Mayo Clin Proc 1996; 71: 117–26.

25. Oudemans-van Straaten HM, Bosman RJ, van der Spoel JL et al. Outcome ofcritically ill patients treated with intermittent high volume hemofiltration: aprospective cohort analysis. Inten Care Med 1999; 25: 814–21.


197

Chap-13.qxd 2/1/02 12:09 PM Page 197

Chap-13.qxd 2/1/02 12:09 PM Page 198


199

14THE ROLE OF THE CARDIOLOGY

CONSULT

This chapter provides a cardiological perspective on the high-risk patient. Mostanaesthetists have extensive experience of anaesthetising patients with cardiac disease.However, cardiologists can be a valuable ally to even the most senior anaesthetist.It is important when requesting a cardiology consult to avoid vague requests forevaluation of fitness for anaesthesia. The cardiologist can be most helpful whenasked specific questions regarding specific issues.

Consultation with a cardiologist may be sought to clarify cardiovascular risk orprognosis prior to a non-cardiac procedure and, consequent to that, perform diag-nostic tests, review or suggest specific management options and provide surveil-lance post-operatively.

PRE-OPERATIVE EVALUATION

In order to identify which patients are most likely to benefit from investigationand treatment, cardiologists have sought to risk stratify patients for peri-operativecardiac complications using a combination of clinical criteria and investigation.

Cardiac surgical risk is a function of

• patient-specific factors,

• surgical-procedure-specific factors.

Patient-specific factors

The basic clinical evaluation obtained by history, physical examination and reviewof the electrocardiograph (ECG) usually provides sufficient data to estimate cardiacrisk. Cardiac risk is defined as the combined incidence of cardiac death and non-fatal myocardial infarction. The history can be used to determine the patient’sfunctional capacity. An assessment of an individual’s capacity to perform a spectrum

Chap-14.qxd 2/1/02 12:09 PM Page 199

of common daily tasks has been shown to correlate well with maximum oxygenuptake by treadmill testing.1

Clinical predictors of increased peri-operative risk for myocardial infarction, heartfailure and death have been established based on multivariate analysis.2–6 Thesefactors have been divided into three categories:

1. Major predictors (recognised markers that mandate intensive managementand may result in delay in surgery):

• acute coronary syndrome,

• decompensated congestive cardiac failure,

• significant arrhythmia,

• severe valvular disease.

2. Intermediate predictors:

• mild angina pectoris,

• prior myocardial infarction either by history or pathological Q-waves,

• compensated or prior history of cardiac failure,

• diabetes mellitus.

3. Minor predictors (recognised markers of cardiovascular disease that have notbeen proven to independently increase peri-operative risk):

• advanced age,

• abnormal ECG,

• rhythm other than sinus, e.g. atrial fibrillation,

• low functional capacity,

• history of stroke,

• uncontrolled systemic hypertension.

FURTHER CARDIAC TESTING PRIOR TO NON-CARDIACSURGERY6

The decision to recommend further non-invasive testing or invasive testing for theindividual patient being considered for surgery ultimately becomes a balancing actbetween the estimated probabilities of effectiveness versus risk; in the process offurther screening and treatment, the risks from the tests and treatments themselvesmay offset or even exceed the potential benefit of evaluation. The proposed benefit,of course, is the possibility of identifying advanced but relatively unsuspected


200

Chap-14.qxd 2/1/02 12:09 PM Page 200

coronary artery disease that might result in significant cardiac morbidity or mor-tality either peri-operatively or long term:

• In general, indications for further cardiac testing and treatments are thesame as in the non-operative setting. The use of both invasive and non-invasive pre-operative testing should be limited to those circumstances in whichthe results of such tests will clearly affect patient management.

• If the surgery is emergent, there may be insufficient time to allow foradequate pre-operative cardiac evaluation. In this setting,post-operativerisk stratification and risk factor management are probably most appropriate.

• In a patient awaiting elective surgery, who has had coronary revascu-larisation within the preceding 5 years and remains asymptomatic and stable, no further testing is generally indicated.

• If the patient has had a favourable recent (within 2 years) coronary evalu-ation and there has been no alteration in symptomatology since then,again cardiac testing is generally not indicated.

• Conversely, in the patient with evidence of an acute coronary syn-drome or one of the other major clinical predictors of risk listed above,elective surgery should be deferred until the problem has been identi-fied and treated.

• Patients with intermediate clinical predictors of clinical risk and mod-erate or excellent functional capacity can undergo intermediate-risksurgery with little likelihood of peri-operative death or myocardialinfarction.

• Non-cardiac surgery is generally safe for patients with neither majornor intermediate predictors of clinical risk and moderate or excellentfunctional capacity.

• Non-invasive testing in combination with the above approach can beused to guide further peri-operative management.

NON-INVASIVE TESTING BEFORE NON-CARDIAC SURGERY

Exercise testing

The test of choice in most ambulatory patients.7–9 Aims:

• to provide an objective measure of functional capacity,

• to identify significant myocardial ischaemia or cardiac arrhythmia,

• to estimate peri-operative cardiac risk and long-term prognosis.

THE ROLE OF THE CARDIOLOGY CONSULT

201

Chap-14.qxd 2/1/02 12:09 PM Page 201

Prognostic gradient of ischaemic responses during ECG-monitored stress test

• High risk – ischaemia induced by low-level exercise (less than 4 METsworkload or heart rate less than 100 beats per minute or less than 70%of age predicted heart rate) MET-metabolic equivalent.

• Intermediate risk – ischaemia induced by moderate-level exercise (4–6METs workload or 70–85% of age predicted heart rate).

• Low risk – no ischaemia or ischaemia induced at high-level exercise(greater than 7 METs workload or heart rate greater than 85% of agepredicted heart rate).

Inadequate stress testInability to reach adequate target workload or heart rate response for age withoutan ischaemic response. In patients with abnormalities of their resting ECG such asleft bundle branch block or unable to perform an adequate stress test, other tech-niques such as dipyridamole thallium testing and dobutamine stress echocardio-graphy should be considered.

Myocardial perfusion imaging methods

Dipyridamole thallium-201 scintigraphy has been extensively used in pre-operative cardiac evaluation.10,11 The negative-predictive value of a normal scan isapproximately 99% for peri-operative myocardial infarction and/or cardiac death.Conversely, those with reversible perfusion defects have an increased risk of devel-oping cardiac complications. However, because the positive-predictive value of anabnormal dipyridamole thallium-201 scan is only between 15% and 30%, otherclinical variables (e.g. previous myocardial infarction, diabetes mellitus, etc.) need to be combined with the results of the perfusion scan to optimise risk stratification.

Dobutamine stress echocardiography

Predictive value of a positive test ranges from 17% to 43%. The negative-predic-tive value ranges from 93% to 100%.12,13 The presence of a new wall motionabnormality is a powerful determinant of an increased risk for peri-operativeevents.

Resting left ventricular function

Evaluated by radionuclide angiography, echocardiography or contrast ventri-culography.14 Resting left ventricular function is not a consistent predictor of peri-operative ischaemic events. In the peri-operative phase, poor left ventricularsystolic or diastolic function is mainly predictive of post-operative congestive heartfailure and, in critically ill patients, death.


202

Chap-14.qxd 2/1/02 12:09 PM Page 202

Ambulatory ECG monitoring

The predictive value of pre-operative ST-changes on 24–48-h ambulatory ECGfor cardiac death or myocardial infarction in patients undergoing vascular andnon-vascular non-cardiac surgery is similar to dipyridamole thallium imaging.15

Ambulatory ECG cannot be performed in a significant percentage of patientsbecause of baseline ECG abnormalities.

CORONARY ANGIOGRAPHY IN PERI-OPERATIVE EVALUATIONBEFORE NON-CARDIAC SURGERY16

Indications

Patients with suspected or proven coronary artery disease with

• high-risk results during non-invasive testing,

• angina pectoris unresponsive to adequate medical therapy,

• most patients with unstable angina pectoris,

• non-diagnostic or equivocal non-invasive testing in high-risk patientundergoing a high-risk non-cardiac surgical procedure.

Not indicated in the following groups:

• Those unwilling to consider a subsequent coronary revascularisationprocedure.

• Severe left ventricular dysfunction and the patient not a candidate forrevascularisation.

• Patient is not a candidate for coronary revascularisation because of co-morbid medical illness.

• Prior technically adequate normal coronary angiogram within 5 years.

• Mild stable angina in a patient with good left ventricular function andlow-risk non-invasive results.

• Asymptomatic patient after coronary revascularisation.

• Low-risk non-cardiac surgery in a patient with known coronary arterydisease and low-risk results on non-invasive testing.

• Screening for coronary artery disease without appropriate non-invasivetesting.


203

Chap-14.qxd 2/1/02 12:09 PM Page 203

MANAGEMENT ASPECTS OF CARDIOVASCULAR CONDITIONS IN THE PERI-OPERATIVE SETTING

Coronary artery disease

Three issues need to be addressed:

• amount of myocardium in jeopardy,

• the amount of stress required to produce ischaemia,

• overall ventricular function.

Clarification of these issues is based on pre-operative history, physical examina-tion and selected non-invasive and invasive testing as outlined above. Dependent on clinical findings and the results of investigation, the patient may require pre-operative coronary revascularisation, optimisation of medical therapy or proceedon to surgery without further delay.

Pre-operative coronary revascularisation

In patients in whom coronary revascularisation is indicated, timing of the proced-ure depends on the urgency of the non-cardiac surgical procedure balancedagainst the stability of the underlying coronary artery disease:

• Patients undergoing elective non-cardiac procedures who are found tohave prognostic high-risk coronary anatomy and in whom long-termoutcome would likely be improved by revascularisation should gener-ally undergo revascularisation before a non-cardiac elective surgical procedure of high or intermediate risk.

• Patients who have previously successfully undergone coronary bypasshave a low peri-operative mortality rate in association with non-cardiacprocedures and the mortality rate is comparable to the surgical risk forpatients who have no clinical indications of coronary artery disease.17,18

Indications for surgical coronary revascularisation:

• Acceptable coronary revascularisation risk and suitable viable myocar-dium with left main stem stenosis.

• Three-vessel coronary artery disease in conjunction with left ventriculardysfunction.

• Two-vessel disease involving severe proximal left anterior descendingartery obstruction.

• Intractable coronary ischaemia despite maximal medical therapy.


204

Chap-14.qxd 2/1/02 12:09 PM Page 204

Pre-operative coronary angioplasty

The role of prophylactic pre-operative coronary angioplasty in reducing untowardperi-operative cardiac complications remains incompletely defined. The indica-tions for angioplasty in the peri-operative setting are currently the same as for theuse of angioplasty in general.19

Following angioplasty, there is uncertainty about what should be the optimal timedelay prior to proceeding on to non-cardiac surgery given the possibility ofrestenosis.

Medical therapy

There are very few randomised trials of medical therapy before non-cardiac surgery to prevent peri-operative cardiac complications. Drugs that have beenconsidered include beta-blockers, nitroglycerin and calcium antagonists.

Beta-blockersBeta-blockers should be used if they have already been required in the recent pastto control symptoms of angina, symptomatic arrhythmias or hypertension. Pre-liminary studies suggest that appropriately administered beta-blockers reduceperi-operative ischaemia and may ultimately reduce risk of myocardial infarctionor death.20

NitroglycerinNitroglycerin has been shown to reverse myocardial ischaemia intra-operatively.High-risk patients previously on nitroglycerin who have active signs of myocardialischaemia without hypotension should receive intra-operative nitroglycerin.21

Calcium antagonistsStudies with diltiazem and nifedipine have been too small to allow for definitiveconclusions to be made on their use in the intra-operative setting.

Hypertension6,22

• Patients with hypertension have higher risks of suffering major cardiaccomplications during or shortly after non-cardiac operation than dopatients who have always been normotensive. Most of the increasedrisk is because of associated coronary artery disease, left ventricular dys-function and renal failure.

• Hypertensive patients are at a greater risk to experience labile bloodpressures and hypertensive episodes during surgery and after extuba-tion. Patients with pre-operative hypertension also appear more likely todevelop intra-operative hypotension than normotensive subjects whichmay be related to a decrease in vascular volume.


205

Chap-14.qxd 2/1/02 12:09 PM Page 205

• Patients with mild to moderate hypertension (diastolic pressures below100 mmHg) and no evidence of serious end-organ damage should beable to tolerate general anaesthesia and major non-cardiac surgery.

• In patients with severe hypertension, particularly of recent onset it isprudent to delay elective surgery to enable evaluation of curable causesof hypertension and obtain control prior to surgery.

• Patients already on antihypertensive therapies should be continued ontheir medication in the peri-operative period. The withdrawal of beta-blockers and clonidine in particular may be associated with reboundeffects on heart rate and/or blood pressure control.

• In emergency surgery rapid-acting agents that allow effective control in a matter of minutes or hours may be required. Beta-blockers are par-ticularly attractive agents in this setting. Several reports have shown thatintroduction of pre-operative beta-blockers leads to effective modu-lation of severe blood pressure fluctuations and a reduction in the number and duration of peri-operative coronary ischaemic complications.

Cardiac failure2,6,14

Congestive heart failure is a major determinant of peri-operative risk, irrespectiveof the nature of the underlying cardiac disorder. The greatest risk of complicationswas observed in patients with a left ventricular ejection fraction of less than 35%:

• Mortality with non-cardiac surgery increases with worsening cardiac classand with the presence of pulmonary congestion.

• The peri-operative mortality is dependent on the patient’s cardiovascu-lar status at the time of operation. Therefore congestive heart failureshould be adequately treated and the patient’s condition stabilised for afew days prior to surgery.

• Hypovolaemia and hypokalemia are potential problems with treatment.Hypovolaemic patients are prone to experience marked hypotensionduring the early phases of anaesthesia.

• Digitalis can counteract the myocardial depression actions of manyanaesthetic agents. It may be of value in patients with congestive heartfailure, who on examination have a third heart sound. Pre-operativedigitalisation because of the risk of intra-operative bradyarrythmias isnot recommended except in patients whose heart failure is sufficientlysevere that they would normally meet the criteria for long-term digitalisation.


206

Chap-14.qxd 2/1/02 12:09 PM Page 206

Post-operative congestive cardiac failureUsually caused by excess fluid administration; may be precipitated by myocardialischaemia or infarction. Tends to occur soon after cessation of positive-pressureventilation.

HYPERTROPHIC CARDIOMYOPATHY23

• Patients are intolerant of hypovolaemia, which may lead to both areduction in the elevated preload needed to maintain cardiac outputand an increase in the obstruction to left ventricular outflow.

• Catecholamines should be avoided because they may increase the degreeof dynamic obstruction and decrease diastolic filling.

• With careful peri-, intra- and post-operative care, the risk of majorcardiac complications is rare. Haemodynamic monitoring may beuseful when these patients undergo major aortic, abdominal or thoracic procedures.

VALVULAR HEART DISEASE24

General points

• Cardiac murmurs are common in patients awaiting non-cardiac sur-gery. Using a combination of examination and echocardiography, thecardiologist will need to determine the underlying aetiology and signifi-cance of any documented murmur.

• Patients with valvular heart disease are subject to many potential hazards:

– heart failure,

– infection,

– arrythmias,

– embolisation.

• Patients with no or only mild limitation of activity tolerate surgery welland probably require little more than careful peri-operative care and prophylaxis for infective endocarditis.

• Those with more serious impairment of cardiac reserve tolerate majornon-cardiac operations poorly, and their prognosis for surviving surgeryis distinctly worse although as is the case for patients with rheumaticheart disease who face the stress of pregnancy, the risk of operationdepends on the functional status of the heart.


207

Chap-14.qxd 2/1/02 12:09 PM Page 207

• Patients with symptomatic critical aortic or mitral stenosis are especially prone to sudden death or acute pulmonary oedema duringthe peri-operative period; this may occur if demands on cardiac outputare suddenly increased or if atrial fibrillation and a rapid ventricular rateare precipitated by anaesthesia or operation.

• Every effort should be made to treat heart failure pre-operatively.

• Patients with severe stenotic or regurgitant valvular disease who requirean emergency operation may benefit from intra-operative haemo-dynamic monitoring, afterload reduction and preload augmentation.

Specific points

• Aortic stenosis – if severe and symptomatic, patients should undergo cor-rective valvular surgery before an elective operation. In rare instances,percutaneous balloon aortic valvuloplasty may be justified when thepatient is not a candidate for valve replacement.

• Mitral stenosis – when stenosis is mild or moderate, adequate control ofthe heart rate must be ensured because the reduction in diastolic fillingthat accompanies tachycardia can lead to severe pulmonary congestion.When the stenosis is severe, the patient may benefit from balloon mitralvalvuloplasty or open surgical repair before high-risk surgery.

• Aortic regurgitation – careful attention to volume control and afterloadreduction is recommended.

• Mitral regurgitation – patients with severe mitral regurgitation benefitfrom afterload reduction and administration of diuretics to producemaximal haemodynamic stabilisation before high-risk surgery.

Prosthetic heart valves24,25

• Most patients with mechanical prosthetic valves are on oral anticoagulantsto prevent thromboembolic complications. If these medications are con-tinued through the period of non-cardiac operation, poor haemostasis,haemotoma formation and persistent post-operative bleeding may ensue.

• In patients who require minimal invasive procedures the INR should bebriefly reduced to the low/sub-therapeutic range with resumption of thenormal dose of anticoagulation immediately following the procedure.

• Peri-operative heparin therapy is recommended for patients in whomthe risk of bleeding on oral anticoagulation is high and the risk ofthromboembolism off anticoagulation is also high.


208

Chap-14.qxd 2/1/02 12:09 PM Page 208

• For patients between these two extremes, physicians must assess the riskand benefit of reduced anticoagulation versus peri-operative heparintherapy.

Endocarditis prophylaxis – patients with valvular heart disease, prosthetic heartvalves and those with congenital heart disease should receive prophylactic antibi-otics for surgical procedures likely to be complicated by bacteraemias.26

CONGENITAL HEART DISEASE

Depending on the nature of the malformation, the patient with congenital heartdisease may be the subject to one or more potentially serious complications:

• Infection.

• Bleeding – patients with cyanotic congenital heart disease and secondarypolycythaemia are at increased risk of intra- and post-operative haem-orrhage as a consequence of coagulation defects and thrombocytopaenia;the risk can be reduced with pre-operative venesection.

• Hypoxaemia.

• Paradoxical embolisation – during general anaesthesia and operation.

ARRHYTHMIAS AND CONDUCTION DEFECTS6

• Cardiac arrhythmias and conduction disturbances are common in theperi-operative period particularly in the elderly.

• The presence of an arrhythmia in the peri-operative setting shouldprompt a thorough search for underlying cardiopulmonary disease,drug toxicity, infection or metabolic derangements.

• Many cardiac arrhythmias are relatively benign. Direct antiarrhythmictherapy is often unnecessary and is usually secondary in importance tocorrection of the underlying cause of the arrhythmia.

• Rarely, arrhythmias because of the haemodynamic or metabolic derange-ments they cause, may deteriorate into more life-threatening rhythmdisturbances.

• Ventricular arrhythmias, whether single premature ventricular contrac-tions, complex ventricular ectopy, or non-sustained ventricular tachy-cardia usually do not require therapy except in the presence of ongoingor threatened myocardial ischaemia or moderate to severe left ventric-ular dysfunction when such arrhythmias represent a significant risk factor.


209

Chap-14.qxd 2/1/02 12:09 PM Page 209

• Drug therapy for supraventricular arrhythmias include digoxin, calciumchannel blockers, beta-blockers and amiodarone. Ventricular arrhyth-mias may respond to intravenous beta-blockers, lidocaine, procainamideor amiodarone.

• Electrical cardioversion should be used for supraventricular or ventric-ular tachyarrhythmias causing haemodynamic compromise.

CONDUCTION DEVICES

Indications for peri-operative temporary/permanent pacing27

• Third degree atrioventricular block associated with the following:symptomatic bradycardia, documented periods of asystole, escaperhythm less than 40 beats per minute in an awake symptom-free patient.

• Second degree atrioventricular block with symptomatic bradycardia.

• Chronic bifascicular and trifascicular block with intermittent thirddegree or Type-II second degree atrioventricular block.

• Sinus node dysfunction with symptomatic bradycardia.

Prophylactic pacemaker placement is not recommended for patients withintraventricular conduction delays, bifascicular block, or left bundle branch blockwith or without first degree atrioventricular block in the absence of a history ofsyncope or more advanced atrioventricular block.

In general, a prophylactic temporary pacemaker should be inserted before non-cardiac operations only if the patient meets the indications for permanent pace-maker insertion and the surgery cannot be delayed for the time required for apermanent pacemaker insertion or the operative course is likely to be complicatedby transient bacteraemia.

The patient with a permanent pacemaker

Permanent pacemakers may need to be checked for end-of-life indicators andprogrammed to verify normal function and the patient’s level of pacemakerdependency. In patients who are totally pacemaker dependent, electrocauteryposes a special problem and should be used only briefly, with the indifferent poleplaced as far away from the pacemaker and heart as possible. In pacemaker-dependent patients, use of bipolar pacing will minimise the risk of electrocautery.

Implanted defibrillators or antitachycardia devices

These devices should be programmed Off immediately before surgery and then On again post-operatively to prevent unwanted discharge due to spurious


210

Chap-14.qxd 2/1/02 12:09 PM Page 210

signals that the device might interpret as ventricular tachycardia or ventricular fibrillation.

POST-OPERATIVE SURVEILLANCE AND THERAPY6

• Intra-aortic balloon counterpulsation device. Placement has been sug-gested as a means of reducing peri-operative cardiac risk but there iscurrently insufficient evidence for its prophylactic use in high-risk non-cardiac surgery.

• Intra- and post-operative use of ST-segment monitoring. Use of computerised ST-segment analysis in appropriate high-risk patientsmay provide increased sensitivity to detect myocardial ischaemia duringthe peri-operative period and may identify patients who benefit fromfurther post-operative intervention.

• Surveillance for peri-operative myocardial infarction. Myocardial infarc-tion occurring in the peri-operative period is often painless. In patientswith known or suspected coronary artery disease undergoing surgicalprocedures associated with a high incidence of cardiovascular morbidity,ECGs at baseline, immediately following surgery, and daily on the first2 days post-operatively appears to be the most cost-effective strategy.Measurement of cardiac enzymes are best reserved for patients at highrisk or those who demonstrate ECG or haemodynamic evidence ofcardiovascular dysfunction.

References

1. Hlatky MA, Boineau RE, Higginbotham MB et al. A brief self-administeredquestionnaire to determine functional capacity (the Duke Activity Status Index).Am J Cardiol 1989; 64: 651–4.

2. Goldman L, Caldera DL, Nussbaum SR et al. Multifactorial index of cardiacrisk in noncardiac surgical procedures. N Engl J Med 1977; 297: 845–50.

3. Detsky AS,Abrams HB,McLaughlin JR et al. Predicting cardiac complicationsin patients undergoing non-cardiac surgery. J Gen Intern Med 1986; 1: 211–19.

4. Mangano DT, Browner WS, Hollenberg M, London MJ,Tubau JF, Tateo IM.Association of perioperative myocardial ischemia with cardiac morbidity andmortality in men undergoing noncardiac surgery: the Study of PerioperativeIschemia Research Group. N Engl J Med 1990; 323: 1781–8.

5. Foster ED, Davis KB, Carpenter JA,Abele S, Fray D. Risk of noncardiac oper-ation in patients with defined coronary disease: the Coronary Artery SurgeryStudy (CASS) registry experience. Ann Thorac Surg 1986; 41: 42–50.


211

Chap-14.qxd 2/1/02 12:09 PM Page 211

6. Report of the American College of Cardiology/American Heart AssociationTask Force on practice guidelines (Committee on Perioperative Cardio-vascular Evaluation for Noncardiac Surgery). Guidelines for perioperativecardiovascular evaluation for noncardiac surgery. J Am Coll Cardiol 1996; 27:910–48; Circulation 1996; 93: 1278–317.

7. Guidelines for exercise testing: a report of the American College ofCardiology/American Heart Association Task Force on assessment of cardio-vascular procedures (Subcommittee on Exercise Testing). Circulation 1997;96 (1): 345–54.

8. Morris CK, Ueshima K, Kawaguchi T, Hideg A, Froelicher VF. The prognos-tic value of exercise capacity: a review of the literature. Am Heart J 1991; 122:1423–31.

9. Chaitman BR. The changing role of the exercise electrocardiogram as a diag-nostic and prognostic test for chronic ischemic heart disease. J Am Coll Cardiol1986; 8: 1195–210.

10. Lette J,Waters D,Cerino M,Picard M,Champagne P,Lapointe J. Preoperativecoronary artery disease risk stratification based on dipyridamole imaging anda simple three-step, three-segment model for patients undergoing noncardiacvascular surgery or major general surgery. Am J Cardiol 1992; 69: 1553–8.

11. Ritchie JL, Bateman TM, Bonow RO et al. Guidelines for clinical use of cardiac radionuclide imaging: a report of the American College of Cardiology/American Heart Association Task Force on assessment of diagnostic and ther-apeutic cardiovascular procedures (Committee on Radionuclide Imaging).J Am Coll Cardiol 1995; 25: 521–47.

12. Lane RT, Sawada SG, Segar DS et al. Dobutamine stress echocardiography for assessment of cardiac risk before noncardiac surgery. Am J Cardiol 1991;68: 976–7.

13. Eichelberger JP, Schwarz KQ, Black ER, Green RM, Ouriel K. Predictivevalue of dobutamine echocardiography just before noncardiac vascular sur-gery. Am J Cardiol 1993; 72: 602–7.

14. Pedersen T, Kelbaek H, Munck O. Cardiopulmonary complications in high-risk surgical patients: the value of preoperative radionuclide cardiography.Acta Anaesthesiol Scand 1990; 34: 183–9.

15. McPhail NV, Ruddy TD, Barber GG, Cole CW, Marois LJ, Gulenchyn KY.Cardiac risk stratification using dipyridamole myocardial perfusion imagingand ambulatory ECG monitoring prior to vascular surgery. Eur J Vasc Surg1993; 7: 151–5.


212

Chap-14.qxd 2/1/02 12:09 PM Page 212

16. Guidelines for coronary angiography: a report of the American College ofCardiology/American Heart Association Task Force on assessment of diagnos-tic and therapeutic cardiovascular procedures (Subcommittee on CoronaryAngiography). Circulation 1999; 99 (17): 2345–57.

17. Reul GJ Jr, Cooley DA, Duncan JM, Frazier OH, Ott DA, Livesay JJ, WalkerWE. The effect of coronary bypass on the outcome of peripheral vascularoperations in 1093 patients. J Vasc Surg 1986; 3: 788–98.

18. Guidelines and indications for coronary artery bypass graft surgery: a report ofthe American College of Cardiology/American Heart Association Task Forceon assessment of diagnostic and therapeutic cardiovascular procedures(Subcommittee on Coronary Artery Bypass Graft Surgery). J Am Coll Cardiol1991; 17: 543–89.

19. Guidelines for percutaneous transluminal coronary angioplasty: a report of the American College of Cardiology/American Heart Association Task Forceon assessment of diagnostic and therapeutic cardiovascular procedures(Committee on Percutaneous Transluminal Coronary Angioplasty). J Am CollCardiol 1993; 22: 2033–54.

20. Pasternack PF, Grossi EA, Baumann FG et al. Beta-blockade to decrease silentmyocardial ischemia during peripheral vascular surgery. Am J Surg 1989; 158:113–16.

21. Coriat P, Daloz M, Bousseau D, Fusciardi J, Echter E, Viars P. Prevention ofintraoperative myocardial ischemia during noncardiac surgery with intra-venous nitroglycerin. Anesthesiology 1984; 61: 193–6.

22. Prys-Roberts C. Hypertension and anesthesia – fifty years on. Anesthesiology1979; 50: 281.

23. Thompson RC, Liberthson RR, Lowenstein E. Perioperative anaesthetic riskof noncardiac surgery in hypertrophic obstructive cardiomyopathy. JAMA1985; 254: 2419–21.

24. Guidelines for the management for the patients with valvular heart disease:a report of the American College of Cardiology/American Heart AssociationTask Force on assessment of diagnostic and therapeutic cardiovascular proced-ures (Committee on the Management of Patients with Valvular Heart Disease).Circulation 1998; 98 (18): 1949–84.

25. Stein PD, Alpert JS, Copeland J, Dalen JE, Goldman S, Turpie AGG.Antithrombotic therapy in patients with mechanical and biological prostheticheart valves. Chest 1992; 102 (suppl.): 445S–55S.

26. AHA medical/scientific statement: prevention of bacterial endocarditis.Circulation 1997; 96 (1): 358–66.


213

Chap-14.qxd 2/1/02 12:09 PM Page 213

27. AHA/ACC guidelines for implantation of pacemakers and antiarrhythmiadevices: a report of the American College of Cardiology/American HeartAssociation Task Force on assessment of diagnostic and therapeutic cardiovas-cular procedures (Committee on Pacemaker Implantation). Circulation 1998;97 (13): 1325–35.


214

Chap-14.qxd 2/1/02 12:09 PM Page 214

215

15THE RISKS OF ANAEMIA AND BLOOD

TRANSFUSION

Anaemia and consequent blood transfusion is relatively common in high-risksurgical and critically ill patients. Only recently, blood transfusion in these patientshas been questioned.

OXYGEN TRANSPORT AND PHYSIOLOGICAL RESPONSE TO ANAEMIA

Whole body oxygen delivery (DO2) is determined by the product of cardiac out-put (CO in l/min) and arterial blood oxygen content (CaO2 in mg/dl):

DO2 � CO � CaO2.

CaO2 is determined primarily by the haemoglobin concentration (Hb in ml/dl)and the degree of Hb oxygen saturation (HbO2/Hb or SaO2, as a fraction), so that

CaO2 � (Hb � SaO2 � K ) � (pO2 � 0.003),

where K is Huffners constant (1.34) – the O2-carrying capacity of 1 g Hb, and pO2

is arterial oxygen tension in mmHg.

It can easily be seen that a fall in Hb may have a profound effect on global DO2

unless compensatory mechanisms occur. It is on this premise that red blood cellsare often transfused, that is, to augment DO2 at a time when the increased cellularoxygen demands of major surgery or critical illness put a strain on the alreadystressed cardiorespiratory systems so that such demands may be met:

• Experimental work suggests an optimal DO2 at an Hb and haematocrit(Hct) of 10 g/dl and 30% respectively, above which the rheologicalproperties of blood cause a reduction in flow and hence a decreasedDO2 overall.

Chap-15.qxd 2/1/02 12:10 PM Page 215

• In the non-critically ill, a drop in Hb concentration results in an increasein erythropoietin (EPO) production within minutes.

• The stimulus to EPO production is a drop in arterial O2 content(CaO2) and so is brought about by both hypoxia and anaemia.

This EPO response appears to be blunted in the critically ill. This bluntedresponse, together with

• decreased iron availability (transferrin saturation � 20% in up to 70% ofpatients),

• direct inhibition of erythropoiesis by the cytokines tumour necrosisfactor and interleukin-1,

• reduced folate level,

contributes to the bone marrow depression typical of critical illness.

In the normovolaemic patient, a rapid drop in Hb brings about certain compen-satory changes:1

• Haemodynamic – the decrease in plasma viscosity improves peripheralblood flow and thus enhances venous return to the right atrium. Animmediate increase in stroke volume follows,by the Starling principle, inresponse to haemodilution and is non-sympathetically mediated. Thereduced viscosity also reduces afterload, which may be an importantmechanism in maintaining CO in the impaired ventricle. Further,increases in CO are mediated through aortic chemoreceptors inducingsympathetically mediated increases in contractility (and so strokevolume), venomotor tone (and thus venous return) and heart rate.

• Microcirculatory – secondary to the increased CO is an increasedorgan capillary blood flow and capillary recruitment. Both of these factors are dependent upon the degree of anaemia and the individualorgan concerned.

• Oxyhaemoglobin dissociation curve (ODC) – a rightward shift in the ODC is seen, which increases the O2 unloading by Hb for agiven blood pO2. This is clearly advantageous in increasing cellular O2

extraction. The primary reason for this is the increased red cell 2,3-diphosphoglycerate (2, 3-DPG) synthesis seen during anaemia. Localtemperature and pH cause a rightward shift in the curve but their effectis thought to be less significant than that of 2,3-DPG.

Note: These are the responses to anaemia. When anaemia is due to acute bloodlosses, the physiological responses to hypovolaemia will also be triggered.


216

Chap-15.qxd 2/1/02 12:10 PM Page 216

ANAEMIA AND THE HEART

Major surgery, critical illness and anaemia all place stress on the myocardium toincrease CO and hence global DO2. To do so, myocardial DO2 must increase tomeet its own increased O2 demand (MVO2). As normal myocardial O2 extractionruns between 75% and 80%, any increase in MVO2 shall be met primarily by anincrease in coronary flow; that is, MVO2 is ‘flow-restricted’. In the presence ofcoronary artery disease, fixed coronary stenoses may prevent any increase inmyocardial flow, thus limiting myocardial DO2. Thus, during anaemia theincreased MVO2 brought about by the demands of an increased CO cannot bemet, coronary blood flow is preferentially diverted to the subepicardial layers andsubendocardial ischaemia or infarction ensue.

PROBLEMS ASSOCIATED WITH BLOOD TRANSFUSION

Problems such as hyperkalaemia, hypocalcaemia, metabolic acidosis, hypothermia,dilutional coagulopathy and citrate toxicity, although important, are related tomassive blood transfusion only and will not be discussed further in this review.

If the purpose of a blood transfusion is to augment tissue oxygen consumption(VO2), then certain so-called ‘storage lesions’ should be borne in mind as theymay have a deleterious effect in this respect:

• Stored blood has reduced levels of 2,3-DPG levels causing a leftwardshift in the ODC and a reduced unloading of O2 from Hb.

• In addition, the reduced membrane deformability of red cells, broughtabout through their storage, is thought to impede their passage throughthe narrow confines of a capillary bed that would otherwise allow thepassage of the more compliant cells unhindered. This may be theexplanation for the observation that patients receiving old transfusedred blood cells developed evidence of splanchnic ischaemia.2

• The high Hct of packed red cells may increase blood viscosity to anextent such that the favourable flow characteristics of anaemia are partially reversed, with a resultant decrease in tissue blood flow and capillary recruitment.

• An interesting study by Purdy3 is the first (and only) study to report anassociation between increased age of transfused red blood cells andoverall mortality in septic patients. The median age of blood unitstransfused to survivors was 17 days versus 25 days for non-survivors.This study was retrospective and therefore one should be cautious inaccepting its implications, but if these findings were confirmed by aprospective randomised trial, it would have major implications for theuse of stored blood in all patients.

THE RISKS OF ANAEMIA AND BLOOD TRANSFUSION

217

Chap-15.qxd 2/1/02 12:10 PM Page 217

Transmission of infection by blood transfusion

There remains an exceedingly small but real risk of infection from transfusion ofstored packed red blood cells:4

• The risk of HIV infection is currently 1 in 4 million units transfused.

• That of hepatitis B is 1 in 100 000–400 000 units.

• Transfusion is becoming an increasingly rare cause of hepatitis C infec-tion, possibly as a result of HIV high-risk donation exclusion togetherwith routine antibody testing of donated blood. The current risk is 1 in200 000 transfusions.

• Infection with Parvovirus B19 is highly variable and appears only to besignificant in pregnancy.

• Bacterial contamination of stored blood is related to the length of stor-age and has a transmission rate of 1 per million transfusions.

• The transmission rates of Plasmodium and Trypanosoma cruzi are vanish-ingly small.

Immunosuppressive effects of blood transfusion5

The immunosuppressive effects of allogenic blood transfusion are well established:

Non-specific Antigen-specific

Reduced natural killer cell production Increased suppressor T-cell and activity production (high CD8 count)

Reduced CD4 helper cell production Anti-idiotypic antibody productionReduced monocyte/macrophage activityReduced interleukin-2 (IL-2) levelsIncreased prostaglandin E2 (PGE2) levels

Postoperative inflammatory responseInitial studies suggested that perioperative allogenic blood transfusion can inducean excessive cytokine response with significant increases in IL-6 in the transfusedgroup. A recent study of cardiac surgery patients6 confirmed that transfusion isassociated with an increased inflammatory response especially IL-6. However, aninteresting finding from this study was that the transfusion packs themselves con-tained increased levels of bactericidal permeability increasing protein (BPI) – amarker of neutrophil activation. Thus, the transfused blood may itself have contained the trigger for part of this inflammatory response. Further, statistical


218

Chap-15.qxd 2/1/02 12:10 PM Page 218

analysis suggested that transfusion was a factor in poor postoperative outcome,although a direct causal relationship was not established.

Postoperative infectionThe effects of blood transfusion-induced immunosuppression have been thought toincrease the risk of postoperative infection.7,8 Much of the evidence is retrospectiveand not all reports have shown an effect of transfusion on postoperative infectionrates. Nonetheless, it remains a possible hazard associated with transfused blood.

It should be remembered that blood transfusion is a common non-infectious causeof leucocytosis in the critically ill.9

Postoperative cancer recurrenceMuch data exists to support the concept of increased tumour recurrence inpatients who have received perioperative allogenic red cell transfusion whilstundergoing potentially curative surgery for cancers.

Vamvakas10 disputes much of the evidence for infection and tumour recurrence,stating that most of the evidence is retrospective and that analysis of prospective,randomised, controlled trials show very little difference between those transfusedand those not transfused. He suggests that differences are due to retrospective trialdesign and that immunomodulation occurs secondary to the variables that lead tothe transfusion and not as a result of the transfusion itself. This is controversial.

Autologous or leucodepleted blood transfusionThe potential evidence for benefit (in terms of reduced complications) fromleucodepleted or autologous blood is as yet inconclusive. Results are variable butmost seem to support either no difference or a reduced incidence of infection andtumour recurrence with leucodepleted and autologous blood.

Blumberg and Heal11 estimate the cost of postoperative transfusion to be on aver-age $1000–2000 per transfusion (blood cost plus complications of transfusion).Spending $240 million on leucodepletion annually in the USA could result insavings of $6–12 billion annually whilst reducing postoperative infections andtumour recurrence rates.

Haemolytic reactionsAn estimated 1 in 250 000–1 000 000 transfusions result in an overt haemolyticreaction, most commonly secondary to minor erythrocyte antigens. This is almostcertainly an underestimate due to under-reporting and failure to recognise such areaction during either surgery or the course of a critical illness at a time when the signs (hypotension, tachycardia, DIC, pyrexia) can be attributed to a more


219

Chap-15.qxd 2/1/02 12:10 PM Page 219

common pathology. The current mortality of 1 per million unit transfusions isattributable to ABO incompatibility as a result of clerical error in 50% of cases.

ROLE OF ANAEMIA AND TRANSFUSION IN SURGICALMORBIDITY

• A retrospective analysis by Carson12 demonstrated a significantly highermorbidity and mortality in surgical patients with a preoperative Hb ofless than 6 g/dl. This effect was substantially more significant in patientswith pre-existing cardiac disease and in those who had a larger bloodloss, as might be expected.

• Other studies have shown an increased cardiac morbidity amongstanaemic surgical patients with cardiovascular disease.13,14

• Carson15 found no increased mortality in 8787 surgical patients downto Hb of 8 g/dl, amongst whom the presence of cardiac disease had nobearing on outcome.

• Speiss et al. demonstrated a negative effect of transfusion on outcome,in patients having coronary artery bypass graft (CABG) surgery.16 Thisretrospective study of 2202 patients investigated the effect of Hct valuesat entry to intensive care unit (ICU) post-surgery on cardiac morbidity.The risk of post-CABG MI was increased when ICU-entry Hct waseither high (� 33%) or medium (25–33%) compared to when ICU-entry Hct was low (� 25%). They also demonstrated a significantlyincreased risk of left ventricular dysfunction and mortality in the highand medium Hct groups compared to the low Hct group.

• In a study of vascular surgery patients, Hb levels of 9 g/dl were toleratedwithout adverse clinical outcome.17 Patients did not compensate foranaemia by increased myocardial work, but by increasing O2 extractionin the peripheral tissues.

• Extreme perioperative haemodilution down to Hb of 5 g/dl inJehovah’s Witnesses has shown no decrease in VO2.

18 In healthy vol-unteers, this degree of anaemia appears to be the point below whichDO2 becomes inadequate (known as the critical point of DO2), andaccumulation of lactic acid occurs as anaerobic metabolism takes over.

Transfusion seems to be commonly triggered by a certain Hb or Hct level with-out any evidence of impaired oxygenation of the tissues. An Hb of 10 g/dl hasbeen traditionally accepted as the level at which Hb should be maintained. Intruth, there is little objective evidence to support this approach and it seems as ifthis number has been chosen for little more than the fact that this is a nice round


220

Chap-15.qxd 2/1/02 12:10 PM Page 220

number! However, evidence already outlined above may support a more liberal‘trigger’ for transfusion in those patients without cardiac disease.

BLOOD TRANSFUSION IN ICU

Studies investigating DO2 in ICU patients suggest an ideal Hct of 33% to aug-ment DO2. Dietrich19 studied the augmentation of DO2 by red cell transfusion involume-resuscitated, non-surgical intensive care patients. He showed neither anincrease in VO2 nor a decrease in blood lactate levels in any patient and concludedthat the shock state of their patient group was not enhanced by red cell transfusion.

Impact of transfusion on outcome in ICU patients

A recent prospective, randomised study20 compared a restrictive with a liberal trans-fusion strategy in 838 general ICU patients:

• Those in the restrictive strategy group were transfused at an Hb � 7 g/dlwith packed red cells to maintain their Hb at 7–9 g/dl.

• Those in the liberal strategy group were transfused with packed redcells at an Hb � 10 g/dl to maintain their Hb at 10–12 g/dl.

Their results are quite striking:

• In those patients with an APACHE score less than 20 and in patientsaged less than 55 years (with APACHE more or less than 20), the 30-daysurvival was significantly better in those randomised to the restrictivetransfusion strategy.

• Amongst patients with cardiac disease, the 30-day mortality was reduced,but not significantly so, if allocated to the restrictive strategy.

• There was a significant decrease in organ dysfunction and cardiac com-plications in the restrictive strategy group.

It is difficult to ignore the implications of such a large and well-conducted study.

Transfusion and weaning from IPPV

• Schonhofer et al.21 demonstrated a decreased work of breathing anddecreased minute ventilation in anaemic patients with chronic obstruct-ive airways disease (COAD) when transfused from an Hb of 8–9 g/dl to an Hb of greater than 12 g/dl. A similar effect was not seen inanaemic patients without COAD who were transfused with an identicalstrategy.

• This work complements case reports by the same authors concerningpatients successfully weaned after transfusion to Hb greater than 12 g/dl.


221

Chap-15.qxd 2/1/02 12:10 PM Page 221

However, ‘difficult to wean’ is a multifactorial problem for which many strategieshave been employed, few with evidence to support them. Transfusion to a normalHb in such patients remains an unproven strategy.

Erythropoietin therapy

• In view of evidence for a blunted EPO response in the critically ill,Corwin et al.22 studied the transfusion requirements of critically illpatients receiving recombinant human erythropoietin (rHuEPO) usingcritically ill patients not receiving rHuEPO as their controls. This ran-domised, controlled, multicentre trial demonstrated a 45% reduction inred cell units transfused with no measured difference in adverse effectsor outcome.

• Other workers have shown an increased P50 amongst cardiac surgicalpatients given rHuEPO.23 This favours well in terms of oxygen trans-port in comparison with the reduced P50 of stored red cells.

EPO therapy is expensive and the benefits are, so far, unproven. However, if itdemonstrably reduces red cell transfusion requirements and avoids the deleteriouseffects of transfusion, it may well be a cost-effective therapy for the critically illpatient.

PRACTICAL GUIDELINES

A major trend in recent years has been greater reluctance to expose patients to theproblems associated with allogenic blood transfusion. This is achieved by:

• Many methods of reducing blood loss during surgery. These are out-side the remit of this review.

• Accepting lesser Hb than previously. In particular, accepting that 10 g/dlis an arbitrary goal in many patients.

Some important principles:

• Recommendations by the ASA Task Force on blood component therapy24 state that transfusion is ‘rarely required above an Hb of 10 g/dl’and transfusion is ‘almost always indicated when Hb is less than 6 g/dl’.

• The multicentre trial in Canadian ICUs would suggest that a restrictivepolicy (i.e. allowing Hbs to drift down to 7–8 g/dl) is safe in ICU patientsand may even be superior (but further studies are needed especially on different subgroups such as neurosurgical patients).

• Many anaesthetists are applying these principles in the operating theatreand this is probably safe and appropriate for most patients. Caution


222

Chap-15.qxd 2/1/02 12:10 PM Page 222

should still, however, be applied in the elderly and those with cardiacdisease. Few now insist that an Hb of 10 g/dl is a prerequisite for elect-ive surgery.

• It would also be prudent not to be too willing to accept Hbs of 7–8 g/dlpreoperatively for those patients whose surgery has the potential forsudden profuse blood loss.

• The ready availability of intraoperative Hb checks is to be encouragedand will permit more accurate control of blood transfusion.

Overall, the majority of practitioners are willing to accept lower Hb in high-risksurgical and other critically ill patients because of

• concerns re-safety of transfused blood,

• evidence that lower Hbs are safe and may even be beneficial,

• lack of convincing evidence of benefits of a liberal transfusion strategy.

Further reading

Supplement on perioperative anaemia. Br J Anaesth 1998; 81: 1–82.

Management of anemia in the critically ill. Crit Care Med 2001; 29: S139–210.

References

1. Hebert PC, Hu L, Biro G. Review of physiological response to anaemia. CanMed Assoc J 1997; 156 (suppl. 11): S27–40.

2. Marik PE, Sibbald WJ. Effect of stored-blood transfusion on oxygen deliveryin patients with sepsis. JAMA 1993; 269: 3024–9.

3. Purdy FR,Tweeddale MG, Merrick PM. Association of mortality with age ofblood transfused in septic ICU patients. Can J Anaesth 1997; 44: 1256–61.

4. British Committee for Standards in Haematology (BCSH). Blood TransfusionTask Force, Guidelines for the clinical use of red cell transfusions. Br JHaematol 2001; 113: 24–31.

5. Landers DF, Hill GE, Wong KC, Fox IJ. Blood transfusion-inducedimmunomodulation. Anesth Analg 1996; 82: 187–204.

6. Fransen E, Maessen J, Dentener M et al. Impact of blood transfusions oninflammatory mediator release in patients undergoing cardiac surgery. Chest1999; 116: 1233–9.

7. Murphy PJ, Connery C, Hicks GL, Blumberg N. Homologous blood transfu-sion as a risk factor for postoperative infection after coronary artery bypassgraft operations. J Thorac Cardiovasc Surg 1992; 104: 1092–9.


223

Chap-15.qxd 2/1/02 12:10 PM Page 223

8. Torchia MG, Danzinger RG. Perioperative blood transfusion and albuminadministration are independent risk factors for the development of postopera-tive infections after colorectal surgery. Can J Surg 2000; 43: 212–16.

9. Fenwick JC, Cameron M, Naiman SC et al. Blood transfusion as a cause ofleucocytosis in critically ill patients. Lancet 1994; 344: 855–6.

10. Vamvakas EC. Transfusion-associated cancer recurrence and postoperativeinfection: meta-analysis of randomised, controlled clinical trials. Transfusion1996; 36: 175–86.

11. Blumberg N, Heal J. Immunomodulation by blood transfusion: an evolvingscientific and clinical challenge. Am J Med 1996; 101: 299–308.

12. Carson JL, Duff A, Poses RM et al. Effect of anaemia and cardiovascular dis-ease on surgical mortality and morbidity. Lancet 1996; 348: 1055–60.

13. Hogue CW, Goodnough LT, Monk TG. Perioperative myocardial ischaemicepisodes are related to haematocrit levels in patients undergoing radical prosta-tectomy. Transfusion 1998; 38: 924–31.

14. Nelson AH, Fleischer LA, Rosenbaum SH. Relationship between postopera-tive anaemia and cardiac morbidity in high-risk vascular patients in the inten-sive care unit. Crit Care Med 1993; 21: 860–6.

15. Carson JL,Duff A,Berlin J et al. Perioperative blood transfusion and postopera-tive mortality. JAMA 1998; 279: 199–205.

16. Speiss BD, Ley C, Body SC et al. Haematocrit value on intensive care unitentry influences the frequency of Q-wave myocardial infarction after coronaryartery bypass grafting. J Thorac Cardiovasc Surg 1998; 116: 460–7.

17. Bush RL, Pevec WC, Holcroft JW. A prospective, randomized trial limitingperioperative red blood cell transfusions in vascular patients. Am J Surg 1997;174: 143–8.

18. Kreimeier U, Messmer K. Haemodilution in clinical surgery; state of the art1996. World J Surg 1996; 20: 1208–17.

19. Dietrich KA, Conrad SA, Hebert CA et al. Cardiovascular and metabolicresponse to red blood cell transfusion in critically ill volume-resuscitated non-surgical patients. Crit Care Med 1990; 18: 940–5.

20. Hebert PC, Wells G, Blajchman MA et al. A multicentre, randomised, con-trolled clinical trial of transfusion requirements in critical care. N Engl J Med1999; 340: 409–17.

21. Schonhofer B, Wenzel M, Geibel M, Kohler D. Blood transfusion and lungfunction in chronically anaemic patients with severe chronic obstructive pulmonary disease. Crit Care Med 1998; 26: 1824–8.


224

Chap-15.qxd 2/1/02 12:10 PM Page 224

22. Corwin HL, Gettinger A, Rodriguez RM et al. Efficacy of recombinanthuman erythropoietin in the critically ill patient: a randomised, double blind,placebo-controlled trial. Crit Care Med 1999; 27: 2346–50.

23. Sowade O, Gross J et al. Evaluation of oxygen availability with the oxygen status algorithm in patients undergoing open heart surgery treated withEpoetin-�. J Lab Clin Med 1997; 129: 97–105.

24. ASA Task Force on blood component therapy. Practice guidelines for bloodcomponent therapy. Anesthesiology 1996; 84: 732–47.


225

Chap-15.qxd 2/1/02 12:10 PM Page 225

Chap-15.qxd 2/1/02 12:10 PM Page 226


16ADMISSION CRITERIA FOR

HDU AND ICU

The provision of intensive care unit (ICU) and high dependency unit (HDU)facilities is an important consideration in patient care in acute service hospitals:

• The recommendation for provision of ICU beds in the UK is 1–2%.

• In 1988 the Association of Anaesthetists of Great Britain and Ireland(AAGBI) identified 55 hospitals with a designated HDU.

• An updated study in 1992/1993 identified only 39 hospitals with anHDU.1

• In 1991, the AAGBI concluded that an intermediate level of acute carebetween that provided on an ICU and general ward would be necessary.

• In the previous year, the Intensive Care Society of Great Britain sug-gested that critical care should be viewed as a spectrum ranging fromthe recovery room, moving through the HDU and culminating in the ICU.

The development of intensive care and high dependency care has been driven bythe perception that severely ill patients may benefit from a greater intensity ofmedical and nursing care than is usually available at general ward level. The weightof clinical opinion supports the view that intensive care improves the survival ofcritically ill patients (or those at risk of becoming critically ill). The distinctionbetween a patient requiring ICU and a patient requiring HDU is complex.

Recent proposals by the Department of Health2 recommend that the existing div-ision into high dependency and intensive care based on beds should be replacedby a classification that focuses on the level of care that individual patients need,regardless of location.

227

Chap-16.qxd 2/1/02 12:10 PM Page 227

They suggest the following levels of care:

Level 0 Patients whose needs can be met through normal ward care in anacute hospital.

Level 1 Patients at risk of their condition deteriorating, or those recentlyrelocated from higher levels of care, whose needs can be met onan acute ward with additional advice and support from the criti-cal care team.

Level 2 Patients requiring more detailed observation or interventionincluding support for a single failing organ system or post-operativecare and those ‘stepping down’ from higher levels of care.

Level 3 Patients requiring advanced respiratory support alone or basic respiratory support together with support of at least two organsystems. This level includes all complex patients requiring sup-port for multi-organ failure.

The review recommends that all acute hospitals carrying out elective surgery mustbe able to provide level 2 care. They should either have level 3 care available onsite or they should have protocols in place to arrange suitable transfer. Hospitalsadmitting emergencies should normally have all levels of care available, although ina limited number of cases, protocols may be agreed for safe transfer to an adjacenthospital for level 3 care.

One of the important uses of an HDU is its role as a ‘halfway house’ between thegeneral ward and the ICU. This means that it would provide a service for the careof patients who have improved after a stay in the ICU but who are not yet wellenough to be transferred back to the ward. Furthermore, an HDU provides a halfway stage for those patients who may subsequently need ICU treatment following clinical deterioration on the general ward. Early intervention and pro-vision of HDU facilities may prevent subsequent ICU admission or favourablyinfluence length of stay in hospital, long-term prognosis, and ultimate survival:

• On a general ward in a hospital without a HDU,up to 13.5% of patientsmay benefit from high dependency care.3

• Many such patients are consequently treated on the ICU, so fuelling thedemand for intensive care services.

• Research also suggests that establishing an intermediate level of carereduces ward mortality rates by 25% and cardiac arrests by 39%.4

The availability of an HDU will thus optimise patient care by

• maintaining the quality of critical care,

• protecting ICU beds for those who need them,


228

Chap-16.qxd 2/1/02 12:10 PM Page 228

• improving the care of patients otherwise treated on general wards,

• improving post-operative pain relief,

• allowing a more cost-effective use of available resources.

DIFFERENCES BETWEEN ICU AND HDU

‘Graduated patient care’ is a concept which allows stratification of patients accord-ing to clinical dependency into those who

• should be admitted to an ICU for the management of single or multipleorgan failure,

• should best be treated in a HDU.

Intensive care appropriate for High dependency care appropriate for

Patients requiring or likely to require Patients requiring support for a single advanced respiratory support alone. failing organ system, but excluding those

needing advanced respiratory support.

Patients requiring support of two or Patients who can benefit from moremore organ systems. detailed observation or monitoring than

can safely be provided on a general ward.

Patients with chronic impairment of Patients no longer needing intensive one or more organ systems sufficient care, but are not yet well enough to beto restrict daily activities and who returned to a general ward.require support of an acute reversible failure of another organ system. Post-operative patients who need close

monitoring for longer than a few hours.

Inappropriate admission

Apart from considerations of cost, inappropriate use of critical care facilities hasother implications.5 The patient may experience unnecessary suffering and loss ofdignity. Relatives may also have to endure considerable emotional pressures. Insome cases, treatment may simply prolong the process of dying or sustain life ofdoubtful quality, and in others the risks of interventions may far outweigh anypotential benefits. To ensure a humane approach to the management of the criticallyill and that limited resources are used appropriately, it is important to identify thosepatients who are most likely to benefit from intensive care and high dependencycare, and to withhold or limit treatment when there is no prospect of recovery.It is also important to avoid admitting those who will make a good recovery without needless iatrogenic intervention.6

ADMISSION CRITERIA FOR HDU AND ICU

229

Chap-16.qxd 2/1/02 12:10 PM Page 229

A patient’s stated or written preference against intensive care should also be takeninto account. Patients or their legal surrogates have the right to control what happens to them. Informed, rational and competent patients therefore have theright to refuse life-sustaining treatment. In addition, patients do not have the rightto demand life-sustaining treatment when the clinician considers it inappropriate.17

As high mortality rates in the ICU and soon after discharge contribute significantlyto the high costs of intensive care, this is a further reason why efforts to containcosts should concentrate on selecting patients with a reasonable chance of survival.7

In practice, however, it is not usually acceptable to deny critically ill patients admis-sion to intensive care, even when there is very little prospect of recovery from theacute illness:

• The long-term prognosis can be unpredictable and it is often difficultto make a precise diagnosis on initial assessment and the only appropriatecourse will be to admit the patient and assess the response after an appro-priate period.

• A recent study comparing the costs of treating survivors with those fornon-survivors estimated that the costs of treating non-survivors wereapproximately three times greater than those of survivors.8

• It is therefore important that the best use is made of intensive careresources and that, as far as possible, the most appropriate group of patientsis admitted.

• Patients who can be expected to receive sustained benefit from intensivecare in terms of quality and length of life should be admitted.

• All decisions should be made jointly by the patient/family, the intensivecare team, and the referring team.

It was previously considered rational to deny admission to the critical care unit onthe grounds of old age. Some elderly patients may have little or no chance of return-ing to an independent existence and there is evidence that age can affect long-termsurvival.9 It appears, though, that physiological age is more important than chrono-logical age in determining survival and that a careful assessment of the patient’s pre-existing physiological reserve and co-morbidity, physical independence and socialcircumstances provides a better indication of the likely benefits of intensive care.10

Limited physiological reserve is known to be an important determinant of mortal-ity11 and intensive care cannot replace lost reserve or reverse chronic ill health.

SUGGESTED ADMISSION CRITERIA FOR HDU AND ICU

Studies looking at the demand for HDUs have used various medical and surgicalcriteria for admission to the HDU and ICU.12,13 The lists below represent theminimum criteria as used in some of these studies. The indications for admission


230

Chap-16.qxd 2/1/02 12:10 PM Page 230

may overlap with services provided by the renal unit, the thoracic HDU and thecoronary care unit.

Medical criteria for admission to HDU

1. Respiratory

(a) Patient requiring more than 40% oxygen on a fixed concentration mask.

(b) Patient with an unstable respiratory condition likely to progress to a deteri-oration classified in (a) or uncontrolled respiratory failure.

(c) Recently extubated patient (within last 6 h). These patients would havebeen discharged from ICU.

(d) Tracheostomy in situ requiring nursing attention more frequently than2 hourly. These patients would have been discharged from ICU.

(e) Intubated patients with no mechanical ventilatory support required through-out day or night.

(f ) Patient requiring the provision of continuous positive airway pressure (CPAP),intubated or unintubated.

(g) Patient with minitracheostomy or similar device in situ requiring attentionincluding physiotherapy at least 2 hourly.

2. Cardiovascular

(a) Patient with a potentially unstable cardiovascular function requiring:(1) continuous ECG monitoring; (2) central venous pressure (CVP) moni-toring; (3) an arterial line in situ for beat to beat pressure monitoring.

(b) Patient with an unstable cardiovascular function or haemodynamic statusdue to haemorrhage – from whatever cause, including post-surgery gastro-intestinal bleeding potentially uncontrolled by peripheral fluid replacement.

(c) Patient with labile blood pressure (due to (b) or any intrinsic cause).

(d) Patients requiring inotropic infusion.

3. Renal

Patients at risk from developing renal failure, e.g. oliguria, in the post-operativeperiod.

4. Pain

Patients whose pain control could not be safely managed in the ward becauseof pre-existing disease.


231

Chap-16.qxd 2/1/02 12:10 PM Page 231

Surgical criteria for admission to HDU

(a) Patient following surgery of an unexpected duration greater than 4 h.

(b) Patient following surgery incorporating unexpected blood loss greater than halfthe circulating blood volume.

Criteria for admission to ICU

(a) After surgery, patients requiring ventilation or who are haemodynamicallyunstable.

(b) Patients requiring mechanical ventilation from any cause.

(c) Treatment of metabolic encephalopathy.

(d) Patients requiring treatment of acute renal failure such as extracorporeal renalreplacement therapy and who cannot be managed on the renal unit.

(e) Patients requiring resuscitation and optimisation before surgery.

Severity scoring indices, the best known of which is APACHE, are useful clinicaltools in predicting probable outcome and in measuring the severity of a patient’scondition. They are valuable for predicting the probability of outcome for acohort of similar patients but not for an individual.14 Kilpatrick et al.15 studiedadmissions to a general ICU and found that 40% of patients were admitted with apredicted risk of mortality of less than 10% using the APACHE score. They sug-gest that patients with a low mortality risk might be better managed on an HDU.Although such scores will continue to be an important adjunct to clinical deci-sion-making and frequently bear out the initial predicted outcome, they are insuf-ficient as a basis for determining admission (or discharge). Indeed, most methodsrely on data collected during the first 24 h of intensive care. In addition, there areinevitably some patients whose death or survival differs from the model’s predic-tion and this invalidates their use as the only basis for deciding on admission (orrefusal) for intensive care or high dependency care.

With regard to surgical patients, the selection of patients most likely to benefit fromadmission to the HDU varies between units and mainly follows local practice.Many of these patients are short stay but require careful observation in the imme-diate post-operative period. Admission criteria relate mainly to the risk of theoperation being performed, the patient’s age, the severity of the patient’s illness, andthe need for close post-operative observation. Obviously, patients of advanced ageundergoing major surgery commonly pose this combination of circumstances. Therisk scoring methods of APACHE II are reliable in the ICU but are not appropri-ate for HDU. POSSUM scores are a very useful indication of operative morbidityand mortality and may well be a better choice in the HDU.16 However, it does notaddress the problem of assessment of surgical patients who do not require surgery.


232

Chap-16.qxd 2/1/02 12:10 PM Page 232

The severity of critical illness and the high expense of providing critical careemphasise the importance of guidelines on admission and discharge. Appropriateadmission guidelines are especially important given that there is some evidenceand opinion to show that patients are admitted to intensive care by necessity in theabsence of other, more appropriate, facilities such as high dependency care:18

• It would seem that medical and nursing opinion is still the most reason-able way to select patients for HDU care.

• The use of selection criteria for ICU admission has not developed sufficiently and is controversial.

• Admission criteria would not be identical between hospitals as manyhospitals have different categories of patients requiring treatment.

• It is equally important to allow admission criteria to evolve to meet theneeds of a changing patient population.

COMPREHENSIVE CRITICAL CARE

Management of surgical patients has become more complex now, as the patientsare older and sicker. Developments in the field of surgery and anaesthetics havelead to more complex therapies and operations:

• These patients are more dependent on pre-operative optimization andpost-operative care.

• At the same time there has been a reduction in the number of nursingstaff in general wards due to recruitment problems.

• Added to this is the reduction of junior doctor working hours, whichhas lead to the shift system.

• There is loss of continuity of care, reduced working relationship betweenjunior doctors and nurses.

All this leads to breakdown of communication that is vital in the care of the highrisk patient.

Thus, we have patients with limited physiological reserve, who are vulnerable tosudden deterioration, which if not promptly treated can lead to catastrophy.

• Recent studies have shown that 40% of ICU admissions are potentiallyavoidable and about 55% of patients admitted to ICU had sub-optimalcare prior to admission to the ICU.

• It was also shown that patients who were appropriately treated in thewards prior to ICU admission fared better.


233

Chap-16.qxd 2/1/02 12:10 PM Page 233

• The difference in ICU mortality between the groups treated appro-priately – 35% and those with sub-optimal care – 56%, is highly significant.19

• The 1993 NCEPOD report showed that two-thirds of peri-operativedeaths occurred 3 or more days after surgery – in the ward,with cardio-pulmonary complications.

This highlights the principle ‘prevention is better than cure’. Numerous studieshave shown that there are premonitory signs and symptoms before the patientsdeteriorate catastrophically.20 The aim should be to identify ‘at risk’ patients,establish appropriate monitoring and develop a system of ‘red flag’ to identifyderangements of physiology so that there can be appropriate corrective measures.

In April 1999 the Department of Health established a review of adult critical careservices. The expert group came up with the idea of ‘comprehensive critical care’,which is a new approach to patients based on the severity of illness. It recom-mends that the doors of intensive care should be opened up to export critical careskills beyond the ICUs to any area within the hospital where such skills areneeded. This is the fundamental concept of critical care outreach. This outreachservice has three main objectives:21

• to avert admission to ICU,

• to enable discharges from ICU,

• to share critical care skills with doctors and nurses in the wards.

The obvious first step, to avert admission to ICU, is for the ward staff looking afterpatients (at level 1 and 2 as discussed previously), to realise who is at risk of becom-ing critically ill. It is difficult to give an exhaustive list of patients who should beconsidered as ‘at risk’, but the following are commonly quoted by many outreachservices as a good guide. They include:

• elderly patients,

• patients admitted as emergencies until stable,

• patients with pre-existing diseases,

• patients whose acute illness is particularly severe,

• patients who fail to progress after treatment,

• patient following ICU discharge,

• patients with decreased level of consciousness,

• patients needing massive blood transfusion.


234

Chap-16.qxd 2/1/02 12:10 PM Page 234

THE EARLY WARNING SCORE FOR SURGICAL PATIENTS22

• The Early Warning Score was developed at the James Paget Hospital,Great Yarmouth, UK to aid the early recognition of patients at risk ofbecoming critically ill.

• It has been developed further at Queens Hospital, Burton-on-Trentresulting in the Modified Early Warning System (MEWS). Urine out-put was added to the early warning score to obtain the modified earlywarning score.

• Essentially the system is a tool, which enables ward-nursing staff tocombine their routine observations to produce an aggregate physiologi-cal score.

• This dynamic scoring system is used to target appropriate help andtreatments to the right patients at the right time.

• If the score remains persistently high then the critical care team mayhave to be involved.

Other similar ‘physiological police’ groups in various countries are: the medicalemergency team; the patient at risk team and the critical care liaison service. Theinput from these teams may prevent an admission to the ICU and when admissionbecomes necessary it will be at an earlier stage of the patients illness.

Generally a MEWS score of � 3 is a trigger to call for senior help. Respiratoryrate scores one or more in 97.5% of the observations in patients who have a totalMEWS score � 4,making it the most sensitive indicator of a deteriorating patient.In comparison, temperature seems to be the least sensitive indicator, scoring � 0in only 7% of the observations (table 16.1).


235

Table 16.1 – The modified early warning score (MEWS).

SSccoorree

33 22 11 00 11 22 33

Respiratory rate � 8 9–14 15–20 21–29 � 30Heart rate � 40 40–50 51–100 101–110 111–129 � 130Blood pressure (%) � 45 30 15 Normal for 15 30 � 45

patientCentral nervous Alert Responds Responds Unresponsivesystem to voice to pain

Temperature � 35.0 35–38.4 � 38.4Urine (ml/kg/h) � 0.5 � 1 � 3

Chap-16.qxd 2/1/02 12:10 PM Page 235

ALERT

The other major aspect of the comprehensive critical care is to take the skills of theintensive care to the wards. To this effect the ALERT (acute life-threatening events –recognition and treatment) course is being implemented in many NHS hospitals.This course, run by a mutidisciplinary team led by intensivists is aimed at nurses ingeneral wards and PRHOs. It is a 1-day course in acute care similar to the ACLSand ATLS, designed specifically to address the high level of sub-optimal ward care.It focuses on the anxieties of ward nurses and PRHOs and the areas of perceivedweakness in the management of acutely ill patients and emphasizes on the recog-nition and early management of sick patients. It sets out a simple assessment andmanagement system that is applicable to everyone.23

MEWS with the recently commissioned ALERT course should be able to identifyat risk patients and provide a quantitative, objective and dynamic indication of thepatient’s status. Like the GCS the MEWS score can be used for better communi-cation between staff. It also helps the nursing staff and junior doctors to pick upthe sick patients at an early stage of their physiological derangement and implementappropriate therapy. This lead time (similar to the golden hour in acute trauma) inthe management of patients should decrease the necessity of ICU/HDU admissionof patients.

References

1. The Royal College of Anaesthetists. National ITU Audit. London: RoyalCollege of Anaesthetists, 1992/1993.

2. Department of Health. Comprehensive Critical Care. A Review of Adult CriticalCare Services. London: HMSO, 2000.

3. Crosby DL, Rees GAD. Provision of postoperative care in UK hospitals. AnnR Coll Surg Engl 1994; 76: 14–18.

4. Franklin CM, Rackow EC, Mandami B et al. Decreases in mortality on a largeurban medical service by facilitating access to critical care. Arch Intern Med1988; 148: 1403–5.

5. Jennett B. Inappropriate use of intensive care. Br Med J 1984; 289: 1709–11.

6. Hinds CJ, Watson D. Intensive Care. A Concise Textbook, 2nd edn, 1996.London: Saunders.

7. Ridley S, Biggam M, Stone P. A cost-utility analysis of intensive therapy.Anaesthesia 1994; 49: 192–6.


236

Chap-16.qxd 2/1/02 12:10 PM Page 236

8. Atkinson A, Bihari D, Sithies M et al. Identification of futility in intensive care.Lancet 1994; 344: 1203–6.

9. Ridley S, Jackson R, Findlay J, Wallace P. Long term survival after intensivecare. Br Med J 1990; 301: 1127–30.

10. Editorial: Intensive care for the elderly. Lancet 1991; 337: 209–10.

11. Bion J. Rationing and triage in intensive care. In Vincent JL (ed.), 1995Yearbook of Intensive Care and Emergency Medicine. Springer Books.

12. Leeson-Payne CG,Aitkenhead AR. A prospective study to assess the demandfor a high dependency unit. Anaesthesia 1995; 50: 383–7.

13. Ryan DW, Bayly PJM, Weldon OGW, Jingree M. A prospective two-monthaudit of the lack of provision of a high-dependency unit and its impact onintensive care. Anaesthesia 1997; 52: 265–75.

14. Teres D, Lemeshow S. Why severity models should be used with caution. CritCare Med 1994; 10: 93–110.

15. Kilpatrick A, Ridley S, Plenderleith L. A changing role for intensive therapy:is there a case for high dependency care? Anaesthesia 1994; 49: 666–70.

16. Jones DR, Copeland GP, de Cossart L. Comparison of POSSUM andAPACHE II for prediction of outcome from a surgical high dependency unit.Br J Surg 1992; 79: 1293–6.

17. Ruark JE, Raffin TA. Initiating and withdrawing life support: principles andpractice in adult medicine. N Engl J Med 1988; 318: 25–30.

18. Metcalfe A,McPherson K. Study of Intensive Care in England 1993,1995. London:HMSO.

19. McQuillan P, Pilkington S, Allan A et al. Confidential inquiry into quality ofcare before admission to intensive care. Br Med J 1998; 316: 1853–8.

20. Franklin C, Matthew J. Developing strategies to prevent in hospital cardiacarrest: analyzing responses of physicians and nurses in the hours before theevent. Crit Care Med 1994; 22: 244–7.

21. Department of Health. Comprehensive Critical Care – Review of Adult CriticalCare Services, 1997.

22. Stenhouse CW,Bion JF. Outreach: a hospital-wide approach to critical illness.Yearbook of Intensive Care and Emergency Medicine, 2001: 661–75.

23. Smith G. ALERT Course Manual, 1st edn, October 2000.


237

Chap-16.qxd 2/1/02 12:10 PM Page 237

Chap-16.qxd 2/1/02 12:10 PM Page 238


239

17THE MEANING OF RISK

• Risk is usually defined as a hazard of loss, or alternatively as the prob-ability of incurring a bad consequence, or misfortune. It is implicitlynegative and is suggestive of a potential danger or hazard and thus isassociated with loss and not gain.

• In 1983 the Royal Society defined risk as ‘the probability that a particu-lar event occurs during a stated period of time or results from a particu-lar challenge’. They defined a hazard as a situation that could lead toharm. The chance or likelihood of this occurring is its associated risk.1

• It is widely recognised that individuals tend to evaluate risks, not solelyon statistical data but on many other subjective qualitative aspects ofrisks. It is also evident that the assessment and perception of risk is subconscious, subjective, personality dependant and fails to follow anyrational or methodical pattern.2

IDENTIFYING RISKS

• Identification of the common potential hazards is not usually a problembut it may be difficult to recognise rare complications particularly withnewly introduced drugs or if there is long lead-time between a treatmentand a complication.

• The timing of any adverse outcomes can have significant effect on theway a particular risk is perceived. Early complications, for example,oftenhave a greater impact than those that are delayed which tend to have adiminished perceived risk value.

• The duration of any adverse outcome can also affect risk perception.Something that is transient like post-operative pain will obviously haveless impact than something more permanent in nature like death or dis-ability. Furthermore, those complications that are easily treated tend tohave downgraded perceived risk severity values.

Chap-17.qxd 2/1/02 12:10 PM Page 239

PERCEIVING RISK

Many previous studies on risk perception have attempted to characterise thoseaspects thought relevant to the way we evaluate risk. The main criteria of risks thatconsciously and subconsciously contribute to the way risks are perceived include:

• magnitude,

• severity,

• vulnerability,

• controllability,

• familiarity,

• acceptability,

• framing effect.

Risk probability or magnitude

• The current accepted method of expressing risk probability or magnitudeof an adverse outcome is in terms of the mathematical probability of anadverse event occurring.

• Estimates of clinical probabilities are usually based on their frequency of occurrence in previously published studies. Risk probabilities quotedneed to be interpreted with caution as accuracy requires large samplesizes, and patient populations studied in other countries may not beapplicable to our own.

• No matter what the actual probability value is, various factors can influ-ence how large, significant or inevitable a risk is perceived to be.

Distortion of the magnitude of risk can be due to two different types of error knownas availability and compression bias.

Availability bias (also known as exposure or publication bias) is an overesti-mation of risk to over exposure or publicity of usually rare, catastrophic or dramaticevents. Probabilities of events are up or downgraded according to the ease withwhich instances of similar events can be recalled:

• Thus rare events are more likely to be sensationalised and are thereforeperceived to be more common than they actually are and conversely,common events are less dramatic, less sensational and therefore under-estimated.

Information availability on a hazard can affect risk perception: for example, wide-spread media coverage of airline crashes increases public anxiety about the risks of


240

Chap-17.qxd 2/1/02 12:10 PM Page 240

airline transport when compared to car travel which is vastly more dangerous interms of fatalities per kilometre travelled.

Compression bias occurs because of the vast ranges that probabilities can span;patients overestimate small risks and underestimate large ones. It is difficult tocommunicate and comprehend rare risks:

• Thus people underestimate the risk of mortality in travelling by bicycleand overestimate the mortality risk of train travel.

Risk severity

This is subjective and perception dependant. The worst outcomes are death ordisability and these obviously have the greatest impact on risk perception.

One mathematical concept used in the past as an attempt to analyse processesinvolved in risk perception,was to compare different risks using expectation value,which is calculated as the product of probability and severity:3

Expectation value � probability � severity.

This is obviously only of use if one can assign a numerical value to severity.

However, it is a considerable oversimplification of the issues we consider whenevaluating a risk for ourselves:

• For example, risks with a very low probability but high severity, forexample death or disability, are perceived worse than risks with a higherprobability and less severe outcome that have the same expectationvalue.

Furthermore, it can be very difficult to assign realistic representative numericalvalues for severity of outcomes that are subjective and perceiver dependant.

Vulnerability

Vulnerability is the extent to which people believe an event could happen to themor alternatively is the degree of immunity one possesses to a risk. Generally wetend to exhibit unrealistic optimism and a feeling of immunity or invincibility sopeople tend not to behave cautiously. Feeling invulnerable, we underestimate ordowngrade our own risk but overestimate the risk to others:

• For example, one might fear more the catastrophic but rare risk ofnuclear accident than the common but minor risk of passive smoking.

Controllability

The possibility of something adverse happening that cannot be controlled magnifies the perceived severity of the risk; we like to be in control; if we can

THE MEANING OF RISK

241

Chap-17.qxd 2/1/02 12:10 PM Page 241

exert some element of control then we feel we can exert influence and minimisethe chance or even prevent the event from occurring. The perception of being incontrol or having choice downgrades the perceived severity of the risk:4

• For example, major risks may be faced regularly (for example, withsmoking or hang-gliding) particularly if individuals deem themselvesinvulnerable risk-takers and perceive that they are in control of the riskswhich they could avoid if they so wished.

• For example, we are often faced with the necessity of travelling from Ato B with certain time constraints forcing the use of a particular modeof public transport offering no other options. Be it flying, rail travel, orthe motorcar most of us accept the risks associated partly throughnecessity and partly through the perception of being in control andexerting some kind of choice.

We are much more willing to accept higher risk levels if they are undertakenvoluntarily than if they are imposed.

On the other hand, involuntary or imposed risks are significantly less acceptable ortolerable:

• For example, risks from passive smoking, or air pollution; the lack ofcontrol incites resentment.

Familiarity

Familiarity of exposure and overconfidence of the extent and accuracy of ourknowledge desensitises us to risks,whereas unfamiliar risks incite a greater degree offear or dread. This distortion is defined in risk terminology as miscalibration bias.

Acceptability or dread

Individual attitudes, upbringing, economic situations, and cultural setting, signifi-cantly affect this concept of fear or dread:

• The loss of a lower limb, for example, might impose greater fear in aprofessional footballer than an office worker.

• More graphically being eaten alive by great white shark usually embodiesfar greater dread than being killed by road traffic accident, even thoughthe final outcome is the same.

The more different characteristics there are embodied in a hazard, the more likelyindividuals’ risk assessments will differ. However likely, severe, controllable, orfamiliar, acceptable the risk seems and however vulnerable or immune the individualfeels will all depend upon a variety of personal experiences and upon the culturalcontext within which the perceiver operates.


242

Chap-17.qxd 2/1/02 12:10 PM Page 242

Framing effect or framing bias

This is how differences in the presentation of risk information can affect perception.

Simply providing risk information on its own is insufficient to change behaviour,but factual information presented effectively can help achieve this:

• In other words, it not what is presented but how risk is presented thatcan have the greatest effect on risk perception and thereby influencebehaviour.

It is well recognised that differences in the presentation of risk information canstrongly affect the perception of risk in both lay people and doctors and therebyinfluence decision making.5 The order in which one chooses to discuss theadvantages or disadvantages of an intervention may have an impact on a patientsperception and final decision and may be one of the many ways in which clinicianscan sway patients final decision on the acceptability of treatments:

• For example, emphasising positive aspects before discussing the risks maybe more likely to persuade an individual to accept a particular therapy.

• Furthermore, adding emphasis to the positive aspects results in a greateruptake; a therapy reported to be 60% effective would be evaluatedmore favourably than by reporting a 40% failure rate, even though thetwo statements are objectively equal.

• Similarly a treatment with 10% mortality will be better received ifphrased as having 90% chance of survival. This is known as positiveframing.5

COMMUNICATING RISK LEVELS

At present there is no universal accepted method for the presentation of probabilityinformation in a format that is readily understood. We have yet to find a formatthat conveys population risk data into clinical risk information that is readilyunderstandable by the individual.6

Because the range of probabilities when expressing risk can be extremely large,and because risk probability data is often only accurate to within an order of magnitude, integer logarithmic scales are often used as a way of presenting riskmagnitude information in a more manageable format.

A number of different integer logarithm based risk scales have been suggested byvarious authors in verbal, numerical and graphical formats:

• Examples of logarithmic scales in everyday use include the Richter scalefor earthquake magnitude, the pH scale for hydrogen ion concentrationand the decibel scale for sound intensity.

THE MEANING OF RISK

243

Chap-17.qxd 2/1/02 12:10 PM Page 243

• All the numerical scales are extremely limited in their use for conveyingrisk magnitude particularly to the layperson; big numbers are simplybeing substituted for smaller numbers with a similar lack of meaning.

• On the other hand Calmans verbal scale2 and his descriptive terms,or thecommunity cluster classification7 are much more useful because of theirvalidity and relevance to the layperson. This is illustrated in table 17.1.

• Others have suggested using the National Lottery and the probabilitiesof the various winning ball combinations as a scale of risk that might bemore understandable to the lay person:8

1 in 57 � 3 balls, 1 in 55 491 � 5 balls, 1 in 13 983 816 � 6 balls,

1 in 1032 � 4 balls, 1 in 2 330 636 � 5 balls � bonus.

WHAT IS HIGH RISK?

Graphical risk ladders have even more impact and meaning when individualexamples of clinical risks are displayed alongside examples of every day risks thatare readily accepted on a daily basis9 (figure 17.1):

• Recently the 1 : 100000 risk level was deemed minimal or even acceptable7

and suggested a risk level of less than 1 : 1 000 000 as being ‘safe’.

• Examples of risks below this ‘acceptable’ frequency of 1 : 100 000 includethe risk of death by murder in 1 year at 1 : 100 000 and the risk of deathby railway accident at 1 : 500 000.

It is enlightening that many of us unwittingly accept the risk of death by road traffic accident in 1 year at 1 in 800010 on our daily journeys to and from work.

• This level of risk below 1 in 1000 is deemed ‘tolerable or reasonable’.2

Some workers however, strongly believe that there is no single level of risk that is universally acceptable.4 For example, some individuals will choose what they


244

Table 17.1 – Easily understood risk scales.

Risk level 1 in … Calmans verbal Calmans descriptive Community clusterscale terms 1 person in a …

1–910–99 High Frequent, significant Family100–999 Moderate Street1000–9 999 Low Tolerable, reasonable Village10 000–99 999 Very low Small town100 000–999 999 Minimal Acceptable Large town1 000 000–9 999 999 Negligible Insignificant safe City

Chap-17.qxd 2/1/02 12:10 PM Page 244

THE MEANING OF RISK

245

Everyday risks Clinical risks1 in 1

1 in 10

1 in 100

1 in 1000

1 in 10 000

1 in 100 000

1 in 1 000 000

1 in 10 000 000

1 in 100 000 000

High

Moderate

Low

Very low

Minimal

Negligible

Very high

Death by murder in 1 year

Death from new variant CJD

Anaesthetic awareness

Neurological injury with spinal

Death all causes to age 40

Death from smoking 10/year

Death by accident at home

Death by accident at work

Death by RTA in 1 year

Death by rail accident

Death from nuclear power accident

Death by lightning strike

6 balls in UK national lottery

Neurological injury with epidural

Death from anaesthesia CEPOD 1982

Maternal deaths from anaesthesiaCEMD 1988–1990

Death from anaesthesia CEPOD 1987

Spinal haematoma after epidural

Spinal haematoma after spinal

Death in 1 year

Figure 17.1 — Risk ladder relating anaesthetic risks to everyday risks (reproduced with permissionfrom Ref. 10).

Chap-17.qxd 2/1/02 12:10 PM Page 245

perceive to be the best alternative for them, and the risk associated with thatchoice must therefore be acceptable to them. In other words, risk magnitude canoften have secondary importance to other subjective criteria involved in theperception of risk.

RISK–BENEFIT ANALYSIS

Risk benefit analysis involves a full assessment of risks and comparing and balan-cing this with the potential gain.

It is a perception dependant process that is particularly reliant on an individual’sanalysis of those advantages and disadvantages of accepting a particular hazard forthe chance of a particular gain.

The mnemonic BRAN offers a useful approach when assessing the risks of acourse of action and includes the Benefits, Risks, Alternatives, and what wouldhappen if you did Nothing!

What are the Benefits?

• Identify the benefits.

• Assess the likelihood of benefit.

• Assess the perceived value of the benefit.

• How soon could benefit occur.

• Is the benefit permanent or temporary.

What are the Risks?

• Identify the risks.

• Assess the likelihood or probability of risk.

• Assess the perceived value of the risk.

• How soon could the risk occur.

• Is the risk permanent or temporary.

What are the Alternatives?

What if you do Nothing?

The BRAN approach may be useful in anaesthetic practice. However, one must know what the risks are before this can be applied to discussions with andmanagement of individual patients.

In the year 2000, the risk of dying in the first 28 days following emergency and non-emergency surgery in the UK was 1 in 25 and 1 in 200, respectively.11


246

Chap-17.qxd 2/1/02 12:10 PM Page 246

Transposing these figures to the risk ladder in figure 17.1 shows that the risks ofundergoing a surgical procedure are not insignificant (table 17.2).

Patient’s (and indeed clinical staff’s) perception of risk is influenced by manyfactors and understanding of probabilities and percentage chances of significantcomplications is poor.

References

1. Royal Society. Risk Assessment: Report of a Royal Society Working Party, 1983.Royal Society, London.

2. Calman KC. Cancer: science and society and the communication of risk.Br Med J 1996; 313: 799–802.

3. Broadbent DE. Psychology of risk. In Cooper MG (ed.) Risk: Man-madeHazards to Man. Oxford: Clarendon Press, 1985.

4. Keeney RL. Understanding life-threatening risks. Risk Anal 1995; 15:627–37.

5. Malenka DJ, Baron JA, Johansen S et al. The framing effect of relative andabsolute risk. J Gen Intern Med 1993; 8: 543–8.

6. Edwards A, Prior L. Communication about risk – dilemmas for general prac-titioners. Br J Gen Prac 1997; 47: 739–42.

7. Calman KC, Royston HD. Risk language and dialects. Br Med J 1997; 315:939–42.

8. Barclay P, Costigan S, Davies M. Lottery can be used to show risk (letter).Br Med J 1998; 316: 124.

9. Adams AM, Smith AF. Risk perception and communication: recent develop-ments and implications for anaesthesia. Anaesthesia 2001; 56: 745–55.

10. BMA Guide to Living with Risk. Harmondsworth: Penguin, 1990.

11. http://www.doh.gov.uk/nhsperformanceindicators.

THE MEANING OF RISK

247

Table 17.2 – Examples of the risks of surgery in the UK.

Statistic Incidence (%) Risk

30-day mortality following cardio-oesophagectomy 10 1 in 1030-day mortality following fractured neck of femur 9.07 1 in 1130-day perioperative mortality – emergency surgery 3.87 1 in 2530-day perioperative mortality – non-emergency surgery 0.48 1 in 200

Chap-17.qxd 2/1/02 12:10 PM Page 247

Chap-17.qxd 2/1/02 12:10 PM Page 248


ABC system, resuscitation 89abdomen

abscesses 175compartment syndrome 183–184pain 59–60sepsis 173surgery

gastrointestinal 165–178respiratory risk 32 (Table)

ABO incompatible blood transfusions,mortality 220

abscesses, percutaneous drainage 175acceptability

risk levels 244treatments 243

accident and emergency departments,patients from 77

acidosis 96–97aortic aneurysm surgery 160

activated neutrophils, reperfusion injuryprevention 162

acute pain services (APS) 51–52techniques 56–57

acute renal failure 179–195renal support 192–194

acute tubular necrosis 180Adamkiewicz, artery of 158adenosine, stress testing 21admission, analgesia 56admission criteria, HDU and ICU 227–237adrenaline

hypertrophic cardiomyopathy 207

local anaesthesia 70, 72–73peri-operative optimisation 122

adrenaline (endogenous), ageing 106adrenal suppression, etomidate 84adult respiratory distress syndrome, ketamine

88aeroplane crashes, availability bias in risk

perception 240–241afterload

aortic regurgitation 145mitral stenosis 147

afterload mismatch, aortic stenosis 143ageing 101–106

by aortic aneurysm surgery 154see also elderly patients

airflow obstruction, value of testing 34airways

aspiration prevention 168elderly patients 103–104, 112management 82–83manipulation and haemodynamics

136patient transfer 78

albumin, ageing 107aldosterone, ageing 105ALERT course 236algorithm controlled opioids 57ambulatory electrocardiographic monitoring

18–19, 24, 203American College of Cardiologists,

ACC/AHA guidelines, pre-operativeassessment 13–15

INDEX

249

Index.qxd 2/2/02 4:47 PM Page 249

American College of Physicians, pulmonaryfunction testing 36

American Heart Association, ACC/AHAguidelines, pulmonary functiontesting 13–15

American Society of AnesthesiologistsASA status classification 9–10, 30, 46task force on blood transfusions 222

aminoglycosides, renal failure 184anaemia 215–255anaesthetic agents 83–88

see also inhalational anaesthetic agents;local anaesthetic agents; specific drugs

anaesthetic rooms 89anaesthetists, competence and clinical risk 2analgesia 51–63

elderly patients, post-operative 113–114thoracic surgery 60

anastomoses, gastrointestinal surgery169–170, 176

anatomical site of surgeryanalgesia 58–61respiratory complications and risk 32–33,

173pain 53

aneurysms, aorta 153–154angiography

coronary 24–25, 130, 203mitral regurgitation 147

angioplasty, coronary 24, 25, 131, 205angiotensin converting enzyme, clinical

risk and 3angiotensin converting enzyme inhibitors

132renal failure 184

angiotensin receptor antagonists 132animal studies

ketamine 87, 88thiopentone 87

antacids 168anterior resection, bowel anastomoses 169antibiotics, renal failure 184anticoagulants

epidural anaesthesia and 54, 61prosthetic heart valves 208–209regional anaesthesia 66–68

antihypertensive drugs 6, 206

antitachycardia devices 210–211aorta, surgery

emergency operations 17, 153–164Goldman’s Cardiac Risk Index 10pulmonary artery catheters 91see also cross-clamping

aortic regurgitation 8, 141, 144–146, 208aortic stenosis 141, 142–144, 208

clinical risk 7–8epidural anaesthesia 60, 144

APACHE scoring systems 13, 232aprotinin, renal failure 184APSs see acute pain servicesarbitrator/coordinators, emergencies 43arbutamine, stress testing 20–21arginine, immune system stimulation 172arrhythmias 209–211

adrenaline 73clinical risk and 8local anaesthetic agents 69valvular heart disease 142

arterial blood gas analysisaortic aneurysm surgery 159pre-operative 37

aortic aneurysm surgery 155arterial lines 90

aortic aneurysm surgery 156arterial oxygen content, anaemia 215arteries

of Adamkiewicz 158ageing 102

arteriovenous fistulae, regional anaesthesia73–74

artificial heart valves 208–209ASA status classification (American Society

of Anesthesiologists) 9–10, 30, 46aspiration risk, gastrointestinal surgery

167–168aspirin

platelet function 67pre-operative 131–132regional anaesthesia 67, 112

asthmaclinical risk 32coronary vasodilators 21

atherosclerosis 108cardiac risk 4

INDEX

250


atracurium, respiratory risk 33atrial fibrillation, mitral stenosis 147atrial natriuretic factor 192

ageing 105atropine, dobutamine stress echocardiography

22audit, NCEPOD on 47–48Austin Flint murmur 145Australian Working Party group (NHMRC),

epidural anaesthesia 54autologous blood transfusions 219

aortic aneurysm surgery 161autonomic ganglia, local anaesthetic agents

69autopsy see post-mortem reviewavailability bias, risk assessment 240–241

bacterial endocarditis prophylaxis 209bacterial infections, blood transfusion risk

218bactericidal permeability increasing protein

218balanced analgesia see multi-modal analgesiaballoon pumps, intra-aortic 133, 211balloons see pulmonary capillary wedge

pressureballoon valvuloplasty 208basal metabolic rate, ageing 106benefits vs risks 246–247�2-adrenoceptors

ageing 102, 105, 106dopexamine 122

�-adrenergic blockersemergency surgery 206hypertension 7intra-operative myocardial ischaemia 137pre-operative 131, 132, 205

biases, risk perception 240–243bicarbonate, rhabdomyolysis 187blood flow

cerebral, ageing 106coronary arteries 134

anaemia 217kidney see renal blood flow

blood gases see arterial blood gas analysisblood loss

epidural anaesthesia 72

hypothermia 81transfusion for 93–94see also haemorrhage

blood pressurewarning of deterioration 235 (Table)see also hypertension; hypotension

blood (product) transfusions 93–94, 217–223aortic aneurysm surgery 159, 161confounding fluid therapy studies 91–92gastrointestinal surgery 170–171haemolytic reactions 219–220NCEPOD recommendations for 43

bone marrow depression 216bowel cancer surgery, surgeons as risk factor

1–2bowel obstruction 175

aspiration risk 167nitrous oxide 168

brachial plexus block 59for arteriovenous fistula 73

bradycardia, ‘paradoxical’ 95BRAN approach, risk-benefit analysis

246–247bronchospasm

coronary vasodilators 21prediction by pulmonary function testing

34surgical risk 32

bupivacaine, epidural anaesthesia 57patient-controlled 58

calcium channel blockers 192pre-operative 131, 132

Calmans verbal and descriptive scales, riskcommunication 244

cancer recurrence, blood transfusions170–171, 219

capacity limited drugs, ageing 107capillaries, anaemia 216cardiac arrest, acute renal failure after 182cardiac enzyme measurement 211cardiac glycosides, digitalis 206cardiac index

peri-operative optimisation 120, 122peri-operative risk 118

cardiac outputageing 102

INDEX

251


cardiac output (contd)anaemia 216aortic aneurysm surgery 160dopamine 191peri-operative risk 118renal failure 185

cardiac risk 4–25, 129, 199–200cardiac surgery

haemofiltration trial 193renal failure 182–183

cardiogenic shock 94cardiology 199–214

see also heartcardiomyopathy 207

epidural anaesthesia 60cardio-oesophagectomy, risk 247 (Table)cardiopulmonary bypass, renal function 187cardiopulmonary reserve, cardiac risk 4carotid endarterectomy, local anaesthesia 73cell savers, blood transfusions 161central nervous system

ageing 105–106dialysis 193local anaesthetic agents 68–69thiopentone 83–84see also neurological injury; neurosurgery

central venous catheters 90aortic aneurysm surgery 156valvular heart disease 142

central venous pressure 90cerebral blood flow, ageing 106Chagas disease, transfusion risk 218Charing Cross protocol, renal failure

prevention 189children, NCEPOD recommendations for

43chloride, metabolic acidosis 97chronic obstructive pulmonary disease

blood transfusions and 221clinical risk 32

Clinical Negligence Scheme for Trusts, ontraining 45

clinical volume, vs risk 1–2clinician-based patient assessment, clinical

risk 17–18clonidine 56, 70–71closing volume (lungs), ageing 103

coagulation defectsaortic aneurysm surgery 159, 161hypothermia 81, 82regional anaesthesia 66–68see also anticoagulants

cocaine 69colloids 91–92

aortic aneurysm surgery 155renal failure 184–185

colon, perforation 176colorectal surgery

Possum score performance 13surgeons as risk factor 1–2

communicationdecisions to operate 46failure 234NCEPOD on 44–45of risk levels 243–244

community-acquired acute renal failure 182community cluster classification, risk

communication 244compartment syndrome, abdomen 183–184compliance (lungs), ageing 103comprehensive critical care 234compression bias, risk perception 241computed tomography, NCEPOD

recommendations for 43computerised ST segment monitoring 137,

211conduction defects 209–211

ageing 102congenital heart disease 209consciousness, level of, warning of

deterioration 235 (Table)consent

elderly patients 112post-mortem examinations 48see also patient refusal

consultantsNCEPOD on 43–44see also senior help requests

consultation, NCEPOD on 44continuous venovenous haemofiltration

192–194contrast media

acute renal failure 182, 184, 187, 188dopamine and 191

INDEX

252


controllability, risk perception 241coordinators (arbitrator/coordinators),

emergencies 43coronary angiography 24–25, 130,

203coronary angioplasty 24, 25, 131, 205coronary arteries

ageing 102–103blood flow 134

anaemia 217occlusion 127–128

see also myocardial infarctionsurgery, anaesthetists’ competence and

clinical risk 2coronary artery bypass grafting 24, 25

blood transfusions, morbidity 220pre-operative 130–131pulmonary artery catheters and 91

coronary artery disease 127–140clinical risk 5–6

predictors 15, 128–129, 200elderly patients 110epidural anaesthesia 54, 137

coronary perfusion pressure 134coronary revascularisation

non-invasive testing and 6pre-operative 130–131, 204recent history of 201, 204see also coronary angioplasty; coronary

artery bypass graftingcoronary vascular resistance 134coronary vasodilators, stress testing 21costs

blood transfusions 219critical care 230

cough, pain on 53creatinine clearance

acute renal failure 180, 181ageing 104dialysis 193

cricoid pressure 168critical care

comprehensive 234outreach 234pre-operative, NCEPOD on 46–47see also high dependency units; intensive

care units

cross-clamping of aorta 158–160renal function 187unclamping 159–160

crystalloids 91–92aortic aneurysm surgery 155renal failure 184–185

cytokines, blood transfusions 218–219

decisions to operate 46defibrillators, implanted 210–211dehydration, elderly patients 110Department of Health, levels of care

227–228desflurane 86dextran

haematoma risk in epidural anaesthesia 68renal failure 184

diabetes mellitus, clinical risk and 8dialysis

intermittent 192–193see also arteriovenous fistulae

diamorphineepidural anaesthesia 57

patient-controlled 58intravenous 56

digitalis 2062,3-diphosphoglycerate

anaemia 216stored blood 217

dipyridamole, stress testing 21, 23, 24, 202diuretics, on kidney 184, 187–188, 190dobutamine

aortic aneurysm surgery 160aortic regurgitation 145–146mitral regurgitation 149pulmonary hypertension 150resuscitation 94stress testing 20–21

echocardiography 21–22, 24, 202dopamine

aortic aneurysm surgery 160renal protection 190–191

dopexamine 122aortic aneurysm surgery 160on kidney 188–189peri-operative optimisation 121, 122

duration of surgery, respiratory risk 33

INDEX

253


Early Warning Score, critical illness 235echocardiography

dobutamine stress testing 21–22, 24,202

left ventricular ejection fraction 23mitral regurgitation 147–148NCEPOD on availability 47see also transoesophageal echocardiography

ejection fractionleft ventricle 23mitral regurgitation 148

elastance, lungs, ageing 103elderly patients 101–116

adrenaline 70cardiac risk 5critical care 230fluid therapy 46, 93, 113general anaesthesia 111–113mortality 42NCEPOD recommendations for 43, 46regional anaesthesia 74, 111, 112renal failure 108, 183respiratory risk 31statistics 109

elective admission, elective surgery,NCEPOD definitions 48

electrocardiographyambulatory monitoring 18–19, 24, 203exercise stress testing 20, 23–24, 130peri-operative 211

electrocautery, pacemakers 210electrolytes, urine, acute renal failure 180emergencies 77–78

admission, NCEPOD definition 48analgesia 56antihypertensive drugs 206cardiac risk management 201coordinators 43coronary artery disease 133fluid therapy, NCEPOD on 46mortality of surgery 247 (Table)operations on aorta 17, 153–164staff availability 45

emergency surgery‘ASA’ status classification 10NCEPOD definition 48risk of 17

emergency theatres, NCEPODrecommendations 43

emphysema, pulmonary function testing36

endocarditis prophylaxis 209endometrial cancer, North American

Negroes 3endothelin antagonists 192enflurane 85, 86enteral nutrition 172Entonox 56, 62enzyme measurement, cardiac enzymes 211epidural anaesthesia 51, 54, 57–58

aortic stenosis 60, 144benefits 71–72complications 61, 66, 73coronary artery disease 54, 137elderly patients 112–113

post-operative care 113–114gastrointestinal surgery 169opioids 57, 60, 70patient-controlled, incident pain 61–62on respiratory complications 173risk levels 245 (Fig.)on stress response to surgery 174–175thorax 60

epinephrine see adrenalineerythropoietin

response 216therapy 222

esmolol, intra-operative myocardial ischaemia137

ethics, critical care admission 229–230ethnicity, on clinical risk 3etomidate 84exercise stress testing 20, 23–24, 130,

201–202exercise tolerance 15–16

exercise stress testing 20respiratory risk 31

expectation value 241expiratory reserve volume, ageing 103exposure bias, risk assessment 240–241

familiarity, risk perception 242fear, risk perception 242felypressin 70

INDEX

254


femoral neck fracturemortality of surgery 247 (Table)peri-operative optimisation 120volume loading 92

fentanyl 85epidural anaesthesia 57, 70

‘fitness for surgery’ 25pulmonary function testing for 35

flow limited drugs, ageing 107fluid balance

aged kidney 104–105elderly patients 110

post-operative 113fluid therapy 91–94

aortic aneurysm surgery 155elderly patients

intra-operative 93post-operative 113pre-operative 46

gastrointestinal surgery 170metabolic acidosis from 97pre-operative, NCEPOD on 46renal failure 187resuscitation 89

forced expiratory volume ‘1’predicted post-operative 36pre-operative 34, 35

forced vital capacity, pre-operative 35fractures

femoral neck see femoral neck fracturemetabolic activity of 96

framing effect, risk perception 243free fractions of drugs, ageing 107frusemide

Charing Cross protocol 189on kidney 187–188

functional capacity see exercise tolerancefunctional residual capacity

ageing 103pain on 52–53

gastrointestinal surgery 165–178gender

cardiac risk 5clinical risk 2–3

genetics, on clinical risk 3gentamicin, renal failure 184

glomerular atrophy, ageing 104glomerular filtration rate, renal failure 181glutamine supplements 172glycosides, digitalis 206Goldman’s Cardiac Risk Index 10–11, 30graduated patient care 229guidelines provision, NCEPOD on 44

haematocritcoronary artery bypass grafting, morbidity

220oxygen delivery 215

haematomaepidural 67spinal, risk levels 245 (Fig.)

haemodialysis, arteriovenous fistulae, regionalanaesthesia 73–74

haemodynamicsacute anaemia 216ageing 106aorta cross-clamping 158–159coronary arteries 134high dependency unit admission criteria

231peri-operative 119–120

haemofiltration 192–194, 195haemoglobin

arterial oxygen content 215levels for transfusion 220–221, 222–223

haemolytic reactions, blood transfusions219–220

haemorrhage 77anaesthesia and 89congenital heart disease 209high dependency unit admission criteria

231see also blood loss

haemorrhagic shock, traumatic shock vs 95–96halothane 85–86hazards, defined 239head injuries 88heart 199–214

ageing 101–103, 108anaemia 216, 217clinical risk 4–25, 118

respiratory risk 30–31epidural anaesthesia 60, 71

INDEX

255


heart (contd)high dependency unit admission criteria

231hypothermia 81inhalational agents on 86ketamine 84, 87local anaesthesia 72–73

agents 69see also cardiac surgery

heart failureclinical risk and 7peri-operative management 206–207

heart ratepulmonary hypertension 150warning of deterioration 235 (Table)

help requests (addressed to seniors) 44, 46heparin

epidural anaesthesia and 61, 67–68prosthetic heart valves 208–209

hepatic clearance of drugs, ageing 107hepatitis B, hepatitis C, blood transfusion

risks 218hepatocellular function, ageing 105high dependency units

admission criteria 227–237advantages 228–229elderly patients 114vs intensive care units 229NCEPOD recommendations 43patients from 77Possum score performance 13

high volume haemofiltration 194, 195hip see femoral neck fracturehistamine receptor antagonists 168HIV infection, blood transfusion risk 218homologous blood transfusions, aortic

aneurysm surgery 161hormones, ageing 106hospital-acquired acute renal failure 182hospitals, facilities required 42–43Huffners constant 215human immunodeficiency virus infection,

blood transfusion risk 218hypercapnia, clinical risk 37hyperoncotic acute renal failure 184hypertension

cardiac risk 6–7

epidural anaesthesia 71haemodynamics 135pre-operative management 205–206

hypertrophic cardiomyopathy 207epidural anaesthesia 60

hyperventilation, raised intracranial pressure 88hypokalaemia, heart failure 206hypotension

acute renal failure 182aorta cross-clamp release 160elderly patients 111epidural anaesthesia 58, 61, 66hypertension patients 205propofol 84

hypothermia 80–82, 111gastrointestinal surgery 167see also warming

hypovolaemiagastrointestinal surgery 170heart failure 206hypertrophic cardiomyopathy 207see also shock

hypoxiaelderly patients 112general anaesthesia 86pain on 53peri-operative 118shock 96

hypoxic reperfusion 162

iatrogenic acute renal failure 182ileus, epidural anaesthesia 72, 169immune system

dopamine on 191nutritional stimulation 172

immunosuppression, blood transfusions218–219

impedance vs resistance, pulmonaryvasculature 150

implanted defibrillators 210–211incentive spirometry 174incident pain 53, 61–62incisions see anatomical site of surgeryinduction of anaesthesia

agents 83–84elderly patients 107–108, 112gastrointestinal surgery 168

INDEX

256


infectionsblood transfusions

from immunosuppression 219transmission 218

epidural anaesthesia, reduction 72gender on clinical risk 2

inflammatory bowel disease 175inflammatory response, blood transfusions

218–219infusions

dopamine 190frusemide 188see also fluid therapy

inguinal blocks 59inhalational anaesthetic agents 85–86

elderly patients 108, 112lactic acidosis 96

inhalational analgesia (Entonox) 56, 62inotropes

aortic aneurysm surgery 160intra-operative 94resuscitation 89stress testing 20–21see also dobutamine; milrinone

integer logarithm risk scales 243–244intensive care units

acute renal failure incidence 182admission criteria 227–237blood transfusions 221–223elderly patients 114high dependency units vs 229NCEPOD recommendations 43patients from 78see also critical care

interleukin-6, blood transfusions 218intermittent dialysis 192–193intermittent positive pressure ventilation

blood transfusions and weaning221–222

gender on clinical risk 3interventricular septum, pulmonary

hypertension 149intestinal obstruction see bowel obstructionintra-aortic balloon pumps 133, 211intracranial haemorrhage 77intracranial pressure increase 88intrapleural local anaesthesia 60

intravenous routedopamine 190frusemide 188heparin, epidural anaesthesia and 67nitrates 137opioids 56see also fluid therapy

intrinsic clearance, ageing 107inulin clearance, ageing 104iron, decreased availability 216ischaemia, aortic aneurysm surgery

161–162spinal cord 158

ischaemic heart disease see coronary arterydisease; myocardial ischaemia

isoflurane 86isoprenaline, pulmonary hypertension 151

jaundice, renal failure risk 183junior staff

critical care problems 233decisions to operate 46

ketamine 84sepsis 88shock 87tissue oxygen extraction 96

ketorolac, limb pain 59kidney

ageing 104–105, 107, 108blood flow see renal blood flowsee also renal failure

lactic acidosis, inhalational anaesthetic agents 96

laparoscopy, respiratory risk 32 (Table)laparotomy, for post-operative sepsis 175laryngoscopy, NCEPOD recommendations

for 43left atrium, mitral stenosis 146–147left ventricle

ageing 102aortic regurgitation 145aortic stenosis 143central venous pressure 90ejection fraction 23end diastolic pressure 135

INDEX

257


left ventricle (contd)radionuclide imaging 22–23resting function 202

leucocytosis, blood transfusions 219leucodepleted blood transfusions 219level of consciousness, warning of

deterioration 235 (Table)levels of care, Department of Health on

227–228limbs, pain 59liver, ageing 105

drug clearance 107liver failure, renal failure 183local anaesthesia 60

emergency admissions 56peripheral nerve blocks 59techniques 65–75

local anaesthetic agents 68–69multi-modal analgesia 55

locum doctors, NCEPOD on 45logarithm risk scales 243–244low molecular weight heparin, epidural

anaesthesia and 67, 68

malaria, blood transfusion risk 218mannitol

on kidney 188on reperfusion injury 162

maximum oxygen uptake 15melanomas, gender on clinical risk 2metabolic acidosis 96–97

aortic aneurysm surgery 160metabolic equivalent levels, exercise

tolerance 15methylxanthines, stress testing and 21metoclopramide 168MEWS (Modified Early Warning Score),

critical illness 235midazolam, shocked patients 87milrinone

aortic regurgitation 145–146mitral regurgitation 149pulmonary hypertension 150

miscalibration bias, risk perception 242mitral regurgitation 141, 147–149, 208mitral stenosis 141, 146–147, 208

clinical risk 7–8

Modified Cardiac Risk Index 11Modified Early Warning Score, critical illness

235monitoring, intra-operative 89, 90–91,

135–136aortic aneurysm surgery 156gastrointestinal surgery 167valvular heart disease 142

morbid obesityanalgesia 59clinical risk 31positive end-expiratory pressure

ventilation 173moribund patients, decision to operate 46morphine 85

algorithm controlled 57gastrointestinal surgery 169intravenous 56

mortality 246–247acute renal failure 194–195anaemia 220aortic aneurysms 153–154, 163blood transfusions 221

ABO incompatible 220packed cells 217

cardiac surgery, renal failure 183on critical care costs 230epidural anaesthesia on 72gastrointestinal surgery 166haemodynamics, peri-operative 119NHS Performance Indicators 42patient profiles 45risk levels 245 (Fig.)

multi-modal analgesia 55–56, 58multiple pain sites 60–61

multiple organ failure syndrome 117–118avoidance 97

murder, risk levels 244murmurs 7muscle relaxants 85

respiratory risk 33myocardial infarction

clinical risk 5ACC/AHA guidelines 15age 5

peri-operative 211prevention 119

INDEX

258


recurrence risk 6 (Table)surgery after 134

myocardial ischaemia 127aortic stenosis 143–144exercise stress testing 20haemodynamic changes causing 135hypothermia 81intra-operative detection and management

137radionuclide imaging 23

myocardiumoxygen management 134–135

anaemia 217perfusion stress radionuclide imaging

22–23, 24, 202revascularisation see coronary

revascularisation

nasogastric tubes 168aortic aneurysm surgery 156

National Confidential Enquiry intoPerioperative Deaths (NCEPOD)41–49

recommendations 42–48statistics on elderly patients 109website 48

National Health Service PerformanceIndicators (NCEPOD) 42

National Lottery, probabilities 244NCCG (non-consultant career grade

doctors), NCEPOD on 45NCEPOD see National Confidential Enquiry

into Perioperative DeathsNegroes, North American

endometrial cancer 3prostate cancer 3

neoplasmsethnicity on risk 3recurrence, blood transfusions 170–171, 219

neostigmine, bowel anastomoses 169neurological injury

anaesthesia, risk levels 245 (Fig.)aortic cross-clamping 158

neurones, ageing 106neuropathic pain 55neurosurgery, NCEPOD recommendations

for 43

neurotransmitters, ageing 106, 107neutrophils, activated 162nicorandil (potassium channel activator),

pre-operative 131, 132nicotine 30night surgery

avoidance 43clinical risk 2

nitrates e.g. nitroglycerinCharing Cross protocol 189intra-operative 205intravenous 137pre-operative 131, 132pulmonary hypertension 150

nitric oxide, pulmonary hypertension 150nitrous oxide 86

gastrointestinal surgery 168non-consultant career grade doctors,

NCEPOD on 45non-invasive testing, coronary artery disease

6, 18–24, 130non-steroidal anti-inflammatory drugs

multi-modal analgesia 55platelet function 67renal failure 183

noradrenaline 94hypertrophic cardiomyopathy 207renal failure prevention 189–190septic shock 95

noradrenaline (endogenous), ageing 106normal pressure hydrocephalus, ageing 105nucleotides, immune system stimulation 172nutritional support, gastrointestinal surgery

171–172

obesityrespiratory risk 31see also morbid obesity

obstructive jaundice, renal failure risk 183oesophagectomy (cardio-oesophagectomy),

risk 247 (Table)oliguria 180

vs non-oliguric renal failure 187–188omega 3 fatty acids, immune system

stimulation 172omeprazole 168operating theatres, management in 77–100

INDEX

259


operations see surgical proceduresopioids

algorithm controlled 57elderly patients 112epidural anaesthesia 57, 60, 70gastrointestinal surgery 169general anaesthesia 84–85multi-modal analgesia 55respiratory depression 54, 70spinal analgesia 70

on stress response 175optimal goals (Shoemaker) 96optimisation, peri-operative 117–125

coronary artery disease 133–134organ donors, colloids and 185organ failure

pain on 53peri-operative optimisation and 123see also multiple organ failure syndrome

orthopaedic surgeryregional anaesthesia 74see also femoral neck fracture

osmolality, urine, acute renal failure 180outreach, critical care 234oxygen consumption

kidney 186, 187peri-operative optimisation 120rewarming 81

oxygen debt 96–97oxygen delivery

anaemia 215critical point 220kidney 186, 187peri-operative optimisation 120, 121,

123oxygen demand, vs supply, myocardium 135oxygen tension, arterial

ageing 103clinical risk 37

oxygen therapyaortic aneurysm surgery 157elderly patients 112

post-operative 113oxygen transport 95–97oxygen uptake, maximum 15oxyhaemoglobin dissociation curve, anaemia

216

pacing, pacemakers (cardiac) 210packed cell transfusions see red blood cell

transfusionspain

ageing on pathways 106assessment 54–55high dependency unit admission criteria

231pathophysiology 52–53

‘Pain after Surgery’ (RCS/RCA) 51–52pancuronium 85

respiratory complications 33, 173paracetamol, multi-modal analgesia 55‘paradoxical’ bradycardia 95parenteral nutrition 171–172parvovirus B19, blood transfusion risk 218patient-controlled analgesia 51, 57

elderly patients 113epidural 58

incident pain 61–62patient positioning 79–80

elderly patients 111patient refusal

epidural anaesthesia 59see also consent

patient selection, peri-operative optimisation122–123

patient transfer 78–79penetrating injuries, permissive hypovolaemia

93perception of risk 239–243percutaneous drainage, abdominal abscesses

175percutaneous transluminal coronary

angioplasty 24, 25, 131, 205perforated viscus 175

colon 176Performance Indicators, National Health

Service, NCEPOD 42perfusion stress radionuclide imaging,

myocardial 22–23, 24, 202peri-operative deaths see National

Confidential Enquiry intoPerioperative Deaths

peri-operative optimisation 117–125coronary artery disease 133–134

peripheral nerve blocks 59

INDEX

260


peripheral vascular disease, clinical risk and8–9

permissive hypovolaemia 93personnel see staffpethidine, gastrointestinal surgery 169pH, intramucosal 170

stomach 4, 92, 123, 170pharmacodynamics, pharmacokinetics,

ageing 107–108physicians, NCEPOD on availability 47‘physiological police’ 235physiotherapy, on respiratory complications

174plane crashes, availability bias in risk

perception 240–241Plasmodium spp., blood transfusion risk 218platelet function, epidural anaesthesia 67positioning see patient positioningpositive end-expiratory pressure ventilation,

morbid obesity 173positive framing 243Possum scoring system 13, 232

NCEPOD on 46post-mortem review, NCEPOD on 47–48post-operative care

aortic surgery 162–163coronary artery disease 137–138, 211elderly patients 113–114heart failure 207, 211high dependency unit admission criteria

232intensive care unit admission criteria

232–233NCEPOD on critical care 47see also analgesia

post-operative forced expiratory volume ‘1’,predicted 36

post-operative hypothermia 80–81potassium channel activator, nicorandil,

pre-operative 131, 132predicted post-operative forced expiratory

volume ‘1’ 36pre-operative assessment

ACC/AHA guidelines 13–15aortic aneurysm surgery 154–155vs clinical risk 25coronary artery disease 128–132

elderly patients 109–110heart 199–203NCEPOD on 45–47respiratory risk 33–37

pressure gradientsaortic stenosis 142mitral stenosis 146

pressure sores, epidural anaesthesia 61probabilities, risk 240

communicating on 243–244prognostic gradient, stress testing 202prokinetic agents, gastrointestinal 168prolactin, dopamine on 191propofol 84

sepsis 87–88prostacycline, pulmonary hypertension 150prostate cancer, North American Negroes 3prosthetic heart valves 208–209protein binding, ageing 107proton pump blockers 168PTCA (coronary angioplasty) 24, 25, 131,

205publication bias, risk assessment 240–241pulmonary artery catheters 90–91, 136, 137

aortic aneurysm surgery 159valvular heart disease 142

mitral regurgitation 149mitral stenosis 147

pulmonary capillary wedge pressure 90peri-operative optimisation 121

pulmonary function testing 34–36pulmonary hypertension 149–151pulmonary volume reduction surgery,

pulmonary function testing 35–36

quality of care, critical care problems233–234

race, on clinical risk 3radiographic contrast media see X-ray

contrast mediaradionuclide myocardial perfusion stress

imaging 22–23, 24, 202railway accidents, risk levels 244rapid sequence induction

aortic aneurysm surgery 157cricoid pressure 168

INDEX

261


red blood cell transfusionsbactericidal permeability increasing

protein 218cancer recurrence 170–171problems 217

reflexes, ageing 106regional anaesthesia 65–66

anticoagulants 66–68aspirin 67, 112coronary artery disease 137elderly patients 74, 111, 112intercurrent disease 72–74stress response to surgery 71, 174–175see also epidural anaesthesia; spinal

anaesthesiaremifentanil 84–85renal blood flow

anaesthesia 185aortic cross-clamping 159vs perfusion pressure 189–190sepsis 186

renal failure 179–197elderly patients 108, 183high dependency unit admission criteria

231local anaesthesia in 73–74

renal insufficiency, definition 179renal replacement therapies 192–194renal tubular function, ageing 104renin, ageing 105reperfusion injury, aortic aneurysm surgery

161–162respiratory centre, local anaesthetic agents 69respiratory depression

epidural anaesthesia 58, 70opioids 54, 70

respiratory rate, warning of deterioration 235respiratory system

ageing 103–104, 108clinical risk, general anaesthesia 33complications 29–39epidural anaesthesia 71gastrointestinal surgery 172–174high dependency unit admission criteria

231local anaesthesia 73

resting function, left ventricle 202

resuscitation 89aortic aneurysm surgery 156

progressive resistance 161–162Revised Cardiac Risk Index 11–13rewarming 81rhabdomyolysis

bicarbonate for 187mannitol 188renal failure 183

rheumatic valve disease 143right ventricle

mitral stenosis 146pulmonary hypertension 149

risk 239–247cardiac 4–25, 129, 199–200of critical illness developing 234factors 1–28gastrointestinal surgery 165–166general anaesthesia 245 (Fig.)indicators (Shoemaker) 118stratification 4–5, 14, 15, 18valvular heart disease 141

risk-benefit analysis 246–247rocuronium 85routes of administration

emergency analgesia 56see also intravenous route; subcutaneous

heparinRoyal College of Anaesthetists

‘Pain after Surgery’ 51–52on supervision 45

Royal College of Surgeons, ‘Pain afterSurgery’ 51–52

Royal Society, definitions of hazard and risk239

saline infusion, metabolic acidosis from 97salvage devices, blood transfusions 161scales

pain assessment 55see also scoring systems

scheduled surgery, NCEPOD definition48

scoring systemsclinical risk 9–13Early Warning Score, critical illness 235see also scales

INDEX

262


senior help requests 44, 46sepsis

abdomen 173acute renal failure 181anaesthetic agents 87–88epidural anaesthesia 60gastrointestinal surgery 175gender on clinical risk 2haemofiltration trial 193inotropes and vasopressors 94–95kidney blood flow 186peri-operative optimisation and 123renal failure 181, 185staff availability on clinical risk 1

septum, interventricular, pulmonaryhypertension 149

sestamibi 22severity of risk 241severity scoring systems

clinical risk 9–13intensive care unit admission criteria 232

sevoflurane 85shock

cardiogenic 94general anaesthesia 86–87haemorrhagic vs traumatic 95–96inotropes and vasopressors 94–95

Shoemaker, W.C., risk indicators 118site of surgery see anatomical site of surgeryskin, patient positioning 80sleep apnoea, analgesia 59smoking, respiratory risk 30sodium

aged kidney 105urine, acute renal failure 180

spinal anaesthesiaaortic stenosis 144benefits 71–72complications 66, 73elderly patients 112–113haematoma risk 68opioids 70

on stress response 175risk levels 245 (Fig.)on stress response to surgery 174–175

spinal cord ischaemia, aortic cross-clamping158

spineelderly patients 112protection 80

splanchnic hypoperfusion 117staff

clinical risk and 1–2critical care problems 233decisions to operate 46NCEPOD recommendations 43–45

sternotomy, pain 60stomach

full, gastrointestinal surgery 167–168intramucosal pH 4, 92, 123, 170

storage lesions (of blood products) 217stress response to surgery

gastrointestinal surgery 174–175regional anaesthesia 71, 174–175

stress testing 19–24, 130exercise 20, 23–24, 130, 201–202prognostic gradients 202

ST segment monitoring, computerised 137,211

subcutaneous heparin, epidural anaesthesiaand 67–68

subendocardial ischaemiaaortic stenosis 143–144radionuclide imaging 23

supervision, NCEPOD on 44, 45supranormal values, peri-operative

optimisation for 120, 121, 123surface area (valve), aortic stenosis 142surgeons, as risk factor 1–2surgical procedures

high dependency unit admission criteria232

risk stratification of 16–17suxamethonium 85sympathetic blockade

epidural anaesthesia 66haemodynamics, elderly patients 106, 113

tachycardiamitral stenosis 147myocardial ischaemia 135pulmonary hypertension 150

technetium-99m, myocardial perfusion stressimaging 22

INDEX

263


temperature, warning of deterioration 235thallium-201, myocardial perfusion stress

imaging 22, 24, 202theatres see emergency theatres; operating

theatresthiopentone 83–84, 87

animal studies 87third heart sound 7third space fluid loss, gastrointestinal surgery

170thoracic cage compliance, ageing 103thoracic surgery

post-operative pain 60respiratory risk 32 (Table)

thromboembolismepidural anaesthesia 71NCEPOD on prophylaxis 47

thyroxine 192tidal volume

ageing 104ventilation, on respiratory complications

173total parenteral nutrition 171–172tracheostomy 82training

ALERT course 236Clinical Negligence Scheme for Trusts on

45need for 1

tramadol, multi-modal analgesia 56transfer of patients 78–79transferrin deficiency 216transfusions see blood (product) transfusionstransoesophageal echocardiography

myocardial ischaemia 137valvular heart disease 142

trauma 77haemofiltration trial 193hypothermia 80peri-operative optimisation 121permissive hypovolaemia 93staff availability on surgical risk 1

traumatic shock, haemorrhagic shock vs95–96

Trypanosoma cruzi, blood transfusion risk 218tubular necrosis, acute 180tubuloglomerular feedback 186

tumoursethnicity on risk 3recurrence, blood transfusions 170–171,

219

United Kingdom, United States, bowelcancer surgery volumes 1–2

uraemia 192urea measurement, renal failure 181urgent admission, urgent surgery, NCEPOD

definitions 48urinary catheterisation, aortic aneurysm

surgery 156urine

electrolytes, acute renal failure 180output, warning of deterioration

235 (Table)timed collections 181

valve surface area, aortic stenosis 142valvular heart disease 141–149, 207–209

clinical risk and 7–8prosthetic valves 208–209

vascular surgeryanaemia, morbidity 220gender on clinical risk 3peri-operative optimisation 120–121renal failure risk 183

vasoconstriction, dopamine 190vasoconstrictors

cocaine 69local anaesthesia 69–70

vasodilatation, rewarming 81vasodilators

coronary, stress testing 21local anaesthetic agents as 69pulmonary hypertension 150

vasopressorsintra-operative 94–95resuscitation 89

vecuronium 85respiratory risk 33

veins, ageing 102venovenous haemofiltration, continuous

192–194ventilation 82–83

blood transfusions and weaning 221–222

INDEX

264


patient transfer 78tidal volume on respiratory complications

173see also intermittent positive pressure

ventilationVIP system, resuscitation 89vital capacity, ageing 103volatile agents see inhalational anaesthetic

agentsvolume (clinical), vs risk 1–2volume loading (fluids)

gastrointestinal surgery 170intra-operative 92–93

vulnerability, risk perception 241

ward carecritical care problems 233

epidural anaesthesia 58patients from 77

warmingaortic aneurysm surgery 156gastrointestinal surgery 167rewarming 81

workload (clinical volume), vs risk 1–2wounds

local anaesthetic infiltration 60, 95metabolic activity of 96

xanthines (methylxanthines), stress testingand 21

X-ray contrast mediaacute renal failure 182, 184, 187, 188dopamine and 191

INDEX

265


anaesthesia for the high risk patient - i. mcconachie (2002) ww

Documents